\\n`````\\n\n",
+ "```\n",
+ "\n",
+ "5 backticks throughout so regular markdown ``` ``` blocks the model emits inside its code/output remain unambiguous. The leading and trailing `\\n` are required — they're part of the marker, so a fence opening at column 0 mid-paragraph still parses cleanly. Result blocks always have lang tag `result` so we can locate them programmatically.\n",
+ "\n",
+ "Rules:\n",
+ "- Use `py` only for Python code and `bash` only for shell commands.\n",
+ "- Do not write a `result` fence yourself; the system will append it after execution.\n",
+ "- After an executable fence, stop immediately. Do not add explanations after the closing fence.\n",
+ "- Use 5 backticks for executable tool fences. Use normal 3-backtick markdown fences for non-executed examples.\n",
+ "- Do not emit a standalone 5-backtick closing line unless it is closing an executable `py` or `bash` fence.\n",
+ "\"\"\""
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "fcc12ec4",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "#| export\n",
+ "def extract_fence_call(text):\n",
+ " \"Return (lang, code) if text ends with terminated py/bash fence, else None\"\n",
+ " ms = list(_fence_re.finditer(text))\n",
+ " if not ms: return None\n",
+ " m = ms[-1]\n",
+ " if m.end() == len(text): return m.group(1), m.group(2)\n",
+ "\n",
+ "def run_fence_tool(lang, code, ns):\n",
+ " \"Run the mapped tool for `lang` with the code, return result fence\"\n",
+ " tname = _lang2tool[lang]\n",
+ " arg = dict(code=code) if lang == 'py' else dict(command=code)\n",
+ " res = call_func(tname, arg, ns=ns, raise_on_err=False)\n",
+ " return _mk_result_fence(_trunc_str(str(res)))"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "a9ece479",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "#| export\n",
+ "class Chat:\n",
+ " def __init__(\n",
+ " self,\n",
+ " model:str, # LiteLLM compatible model name \n",
+ " sp='', # System prompt\n",
+ " temp=0, # Temperature\n",
+ " search=False, # Search (l,m,h), if model supports it\n",
" tools:list=None, # Add tools\n",
" hist:list=None, # Chat history\n",
" ns:Optional[dict]=None, # Custom namespace for tool calling \n",
@@ -3635,23 +4215,24 @@
" if completefunc is None: completefunc = acompletion if any(o.__name__=='AsyncChat' for o in type(self).mro()) else completion\n",
" store_attr()\n",
" \n",
- " def _prep_msg(self, msg=None, prefill=None):\n",
+ " def _prep_msg(self, msg=None, prefill=None, stop_callables=None):\n",
" \"Prepare the messages list for the API call\"\n",
" sp = [{\"role\": \"system\", \"content\": self.sp}] if self.sp else []\n",
" if sp:\n",
" if 0 in self.cache_idxs: sp[0] = _add_cache_control(sp[0])\n",
" cache_idxs = L(self.cache_idxs).filter().map(lambda o: o-1 if o>0 else o)\n",
" else: cache_idxs = self.cache_idxs\n",
- " if msg: self.hist = self.hist+[msg]\n",
- " self.hist = mk_msgs(self.hist, self.cache and 'claude' in self.model, cache_idxs, self.ttl)\n",
+ " if msg: self.hist = self.hist + [msg]\n",
+ " msgs = mk_msgs(self.hist, self.cache and 'claude' in self.model, cache_idxs, self.ttl)\n",
" pf = [{\"role\":\"assistant\",\"content\":prefill}] if prefill else []\n",
- " msgs = sp + self.hist + pf\n",
+ " msgs = sp + msgs + pf\n",
+ " msgs = _split_fence_msgs(msgs) if stop_callables else msgs\n",
" if self.tool_reminder: msgs = _inject_tool_reminder(msgs, self.tool_reminder)\n",
" if 'deepseek' in self.model:\n",
" for m in msgs:\n",
" if m.get('role')=='assistant' and not m.get('reasoning_content'): m['reasoning_content'] = ''\n",
" return msgs\n",
- " \n",
+ "\n",
" @property\n",
" def tcdict(self): return dict(tool_schemas=self.tool_schemas, ns=self.ns, tc_res=self.tc_res, tc_res_eval=self.tc_res_eval)\n",
" def _track(self, res):\n",
@@ -3730,6 +4311,21 @@
" return dict(reasoning_effort=eff)"
]
},
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "95986c76",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "#| export\n",
+ "def _active_fence_langs(tool_schemas):\n",
+ " \"Return set of active fence langs whose mapped tool is registered\"\n",
+ " if not tool_schemas: return set()\n",
+ " names = {nested_idx(t, 'function', 'name') for t in tool_schemas}\n",
+ " return {lang for lang, tname in _lang2tool.items() if tname in names}"
+ ]
+ },
{
"cell_type": "code",
"execution_count": null,
@@ -3762,6 +4358,9 @@
" if self.extra_headers: kwargs['extra_headers'] = self.extra_headers\n",
" if stream: kwargs['stream_options'] = {\"include_usage\": True}\n",
" kwargs.update(_think_kw(self.model, think))\n",
+ " if (langs := _active_fence_langs(self.tool_schemas)):\n",
+ " if not any(isinstance(s, FenceToolStop) for s in kwargs.get('stop_callables', [])):\n",
+ " kwargs['stop_callables'] = kwargs.get('stop_callables', []) + [FenceToolStop(langs)]\n",
" return prefill, max_tokens"
]
},
@@ -3775,19 +4374,20 @@
"#| export\n",
"@patch\n",
"def _call(self:Chat, msg=None, prefill=None, temp=None, think=None, search=None, stream=False,\n",
- " max_steps=2, step=1, final_prompt=None, tool_choice=None, max_tokens=None, **kwargs):\n",
+ " max_steps=2, step=1, final_prompt=None, tool_choice=None, max_tokens=None, stop_callables=None, **kwargs):\n",
" \"Internal method that always yields responses\"\n",
" if step>max_steps+1: return\n",
" if self.callkw: kwargs = {**self.callkw, **kwargs}\n",
" prefill, max_tokens = self._prep_call(prefill, search, max_tokens, kwargs, stream=stream, think=think)\n",
+ " stop_callables = listify(stop_callables) + listify(kwargs.pop('stop_callables', None))\n",
" mt = {} if max_tokens in (None,0) else dict(max_tokens=int(max_tokens))\n",
- " res = self.completefunc(model=self.model, messages=self._prep_msg(msg, prefill), stream=stream, **mt,\n",
+ " res = self.completefunc(model=self.model, messages=self._prep_msg(msg, prefill, stop_callables), stream=stream, **mt,\n",
" tools=self.tool_schemas, tool_choice=tool_choice, num_retries=2,\n",
" temperature=None if think else ifnone(temp,self.temp),\n",
" caching=self.cache and 'claude' not in self.model, **kwargs)\n",
" if stream:\n",
" if prefill: yield _mk_prefill(prefill)\n",
- " res = yield from stream_with_complete(res, postproc=cite_footnotes)\n",
+ " res = yield from stream_with_complete(res, postproc=cite_footnote, stop_callables=stop_callables)\n",
" elif not isinstance(res, ModelResponse): res = stream_chunk_builder(list(res))\n",
" m = contents(res)\n",
" if prefill: m.content = prefill + m.content\n",
@@ -3814,7 +4414,19 @@
" except ContextWindowExceededError:\n",
" for t in tool_results:\n",
" if len(t['content'])>1000: t['content'] = _cwe_msg + _trunc_str(t['content'], mx=1000)\n",
- " yield from self._call(None, prefill, temp, think, search, stream, max_steps, max_steps, final_prompt, 'none', **kwargs)"
+ " yield from self._call(None, prefill, temp, think, search, stream, max_steps, max_steps, final_prompt, 'none', **kwargs)\n",
+ " elif (langs := _active_fence_langs(self.tool_schemas)):\n",
+ " m = self.hist[-1]\n",
+ " if m.role == 'assistant':\n",
+ " if fence := extract_fence_call(m.content or ''):\n",
+ " lang, code = fence\n",
+ " out = run_fence_tool(lang, code, self.ns)\n",
+ " m.content += out \n",
+ " if stream: yield mk_stream_chunk(content=out, role='assistant')\n",
+ " if step <= max_steps:\n",
+ " yield from self._call(\n",
+ " None, prefill, temp, think, search, stream, max_steps, step+1,\n",
+ " final_prompt, tool_choice, **kwargs)"
]
},
{
@@ -3863,147 +4475,99 @@
},
{
"cell_type": "markdown",
- "id": "ce163563",
+ "id": "26f75663",
"metadata": {},
"source": [
- "## Examples"
+ "## Streaming Display"
]
},
{
"cell_type": "markdown",
- "id": "1dbf319a",
+ "id": "922690a2",
"metadata": {},
"source": [
- "### History tracking"
+ "This is what our outputs look like with streaming results:"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "9944d011",
+ "id": "35b176d0",
"metadata": {},
"outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Sure! Let me calculate that for you right away.\n",
+ "🔧 simple_add\n"
+ ]
+ },
{
"data": {
- "text/markdown": [
- "Your name is Rens.\n",
- "\n",
- "\n",
- "\n",
- "- id: `chatcmpl-xxx`\n",
- "- model: `gpt-5.4-2026-03-05`\n",
- "- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=9, prompt_tokens=40, total_tokens=49, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=0, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None))`\n",
- "\n",
- " "
- ],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gpt-5.4-2026-03-05', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='Your name is Rens.', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'refusal': None}, annotations=[]), provider_specific_fields={})], usage=Usage(completion_tokens=9, prompt_tokens=40, total_tokens=49, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=0, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None)), service_tier='default')"
+ "{'tool_call_id': 'toolu_016d4zS8Gz82ZkEJSWVqjzkU',\n",
+ " 'role': 'tool',\n",
+ " 'name': 'simple_add',\n",
+ " 'content': '12'}"
]
},
- "execution_count": null,
"metadata": {},
- "output_type": "execute_result"
+ "output_type": "display_data"
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "The result of **5 + 7 = 12**. 🎉"
+ ]
}
],
"source": [
- "for m in ms[1:]:\n",
- " chat = Chat(m)\n",
- " chat(\"Hey my name is Rens\")\n",
- " r = chat(\"Whats my name\")\n",
- " test_eq('Rens' in contents(r).content, True)\n",
- "r"
+ "chat_with_tools = Chat(model, tools=[simple_add])\n",
+ "res = chat_with_tools(\"What is 5 + 7? Use the tool to calculate it.\", stream=True)\n",
+ "for o in res:\n",
+ " if isinstance(o,ModelResponseStream): print(delta_text(o) or '',end='')\n",
+ " elif isinstance(o,dict): _display(o)"
]
},
{
"cell_type": "markdown",
- "id": "8dbb77c7",
- "metadata": {},
- "source": [
- "See now we keep track of history!\n",
- "\n",
- "History is stored in the `hist` attribute:"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "4a3c29d8",
+ "id": "6826291d",
"metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "[{'role': 'user', 'content': 'Hey my name is Rens'},\n",
- " Message(content='Hi Rens! Nice to meet you — how can I help today?', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'refusal': None}, annotations=[]),\n",
- " {'role': 'user', 'content': 'Whats my name'},\n",
- " Message(content='Your name is Rens.', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'refusal': None}, annotations=[])]"
- ]
- },
- "execution_count": null,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
"source": [
- "chat.hist"
+ "Here's a complete `ModelResponse` taken from the response stream:"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "3d010ee1",
+ "id": "3da6eb05",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
- "{'role': 'user', 'content': 'Hey my name is Rens'}\n",
- "\n",
- "Message(content='Hi Rens! Nice to meet you — how can I help today?', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'refusal': None}, annotations=[])\n",
- "\n",
- "{'role': 'user', 'content': 'Whats my name'}\n",
- "\n",
- "Message(content='Your name is Rens.', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'refusal': None}, annotations=[])\n",
- "\n"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-5', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=\"I'll calculate ((10 + 5) * 3) / (2 + 1) step by step:\", role='assistant', tool_calls=[ChatCompletionMessageToolCall(function=Function(arguments='{\"a\": 10, \"b\": 5}', name='simple_add'), id='toolu_018BGyenjiRkDQFU1jWP6qRo', type='function'), ChatCompletionMessageToolCall(function=Function(arguments='{\"a\": 2, \"b\": 1}', name='simple_add'), id='toolu_01CWqrNQvoRjf1Q1GLpTUgQR', type='function')], function_call=None, provider_specific_fields=None))], usage=Usage(completion_tokens=228, prompt_tokens=794, total_tokens=1022, completion_tokens_details=None, prompt_tokens_details=None))\n"
]
}
],
"source": [
- "chat.print_hist()"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "8f38015b",
- "metadata": {},
- "source": [
- "You can also pass an old chat history into new Chat objects:"
+ "resp = ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-5', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=\"I'll calculate ((10 + 5) * 3) / (2 + 1) step by step:\", role='assistant', tool_calls=[ChatCompletionMessageToolCall(function=Function(arguments='{\"a\": 10, \"b\": 5}', name='simple_add'), id='toolu_018BGyenjiRkDQFU1jWP6qRo', type='function'), ChatCompletionMessageToolCall(function=Function(arguments='{\"a\": 2, \"b\": 1}', name='simple_add'), id='toolu_01CWqrNQvoRjf1Q1GLpTUgQR', type='function')], function_call=None, provider_specific_fields=None))], usage=Usage(completion_tokens=228, prompt_tokens=794, total_tokens=1022, prompt_tokens_details=None))\n",
+ "print(repr(resp))"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "d9f575f3",
+ "id": "fb2a4cc1",
"metadata": {},
"outputs": [
{
"data": {
- "text/markdown": [
- "Your name is Rens.\n",
- "\n",
- "\n",
- "\n",
- "- id: `chatcmpl-xxx`\n",
- "- model: `gpt-5.4-2026-03-05`\n",
- "- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=9, prompt_tokens=62, total_tokens=71, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=0, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None))`\n",
- "\n",
- " "
- ],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gpt-5.4-2026-03-05', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='Your name is Rens.', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'refusal': None}, annotations=[]), provider_specific_fields={})], usage=Usage(completion_tokens=9, prompt_tokens=62, total_tokens=71, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=0, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None)), service_tier='default')"
+ "ChatCompletionMessageToolCall(function=Function(arguments='{\"a\": 10, \"b\": 5}', name='simple_add'), id='toolu_018BGyenjiRkDQFU1jWP6qRo', type='function')"
]
},
"execution_count": null,
@@ -4012,263 +4576,266 @@
}
],
"source": [
- "for m in ms[1:]:\n",
- " chat2 = Chat(m, hist=chat.hist)\n",
- " r = chat2(\"What was my name again?\")\n",
- " test_eq('Rens' in contents(r).content, True)\n",
- "r"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "fad4b9d6",
- "metadata": {},
- "source": [
- "If max tokens limit is reached, a custom warning message will be added to the end of the model response:"
+ "tc=resp.choices[0].message.tool_calls[0]\n",
+ "tc"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "331923bf",
+ "id": "abf0c460",
"metadata": {},
- "outputs": [
- {
- "data": {
- "text/markdown": [
- "Every morning at exactly 7:03, Unit 7 rolled out of the garage and checked the front gate, the mailbox, and the tomato plants. It was very good at routines\n",
- "\n",
- "Response was cut off at token limit. \n",
- "\n",
- "\n",
- "\n",
- "- id: `chatcmpl-xxx`\n",
- "- model: `gpt-5.4-2026-03-05`\n",
- "- finish_reason: `length`\n",
- "- usage: `Usage(completion_tokens=40, prompt_tokens=16, total_tokens=56, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=0, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None))`\n",
- "\n",
- " "
- ],
- "text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gpt-5.4-2026-03-05', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='length', index=0, message=Message(content='Every morning at exactly 7:03, Unit 7 rolled out of the garage and checked the front gate, the mailbox, and the tomato plants. It was very good at routines\\n\\nResponse was cut off at token limit. ', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'refusal': None}, annotations=[]), provider_specific_fields={})], usage=Usage(completion_tokens=40, prompt_tokens=16, total_tokens=56, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=0, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None)), service_tier='default')"
- ]
- },
- "execution_count": null,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
+ "outputs": [],
"source": [
- "chat_long = Chat(m)\n",
- "r = chat_long(\"Write a short story about a robot and a dog\", max_tokens=40)\n",
- "r"
+ "tr={'tool_call_id': 'toolu_018BGyenjiRkDQFU1jWP6qRo', 'role': 'tool','name': 'simple_add',\n",
+ " 'content': '15 is the answer! ' +'.'*2000}"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "5447f11a",
+ "id": "0d4e9e9a",
"metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Every morning at exactly 7:03, Unit 7 rolled out of the garage and checked the front gate, the mailbox, and the tomato plants. It was very good at routines\n",
- "\n",
- "Response was cut off at token limit. \n"
- ]
- }
- ],
+ "outputs": [],
"source": [
- "print(contents(r).content)"
+ "#| export\n",
+ "def trunc_param(v, mx=40):\n",
+ " \"Truncate and escape param value for display\"\n",
+ " tp = _trunc_str(str(v).replace('`', r'\\`'), mx=mx, replace=None, skip=0)\n",
+ " try: return ast.literal_eval(tp)\n",
+ " except Exception: return repr(tp).replace('\\\\\\\\', '\\\\')\n",
+ "\n",
+ "def _tc_summary(tc, tr=None):\n",
+ " \"Format tool call as func(params) → result string\"\n",
+ " args = json.loads(tc.function.arguments)\n",
+ " params = ', '.join(f\"{k}={trunc_param(v)}\" for k,v in args.items())\n",
+ " res = f\"→{trunc_param(tr.get('content',''))}\" if tr else ''\n",
+ " return ''+escape(f\"{tc.function.name}({params}){res}\")+''\n",
+ "\n",
+ "def _trunc_content(content, mx):\n",
+ " \"Truncate tool result content, respecting '_full' flag\"\n",
+ " if isinstance(content, dict) and '_full' in content and len(content)==1: return content['_full']\n",
+ " return _trunc_str(content, mx=mx)\n",
+ "\n",
+ "def mk_tr_details(tr, tc, mx=2000):\n",
+ " \"Create block for tool call as JSON\"\n",
+ " args = {k:_trunc_str(v, mx=mx*5) for k,v in json.loads(tc.function.arguments).items()}\n",
+ " res = {'id':tr['tool_call_id'],\n",
+ " 'call':{'function': tc.function.name, 'arguments': args},\n",
+ " 'result':_trunc_content(tr.get('content'), mx=mx),}\n",
+ " summ = f\"{_tc_summary(tc,tr)} \"\n",
+ " return f\"\\n\\n{tool_dtls_tag}\\n{summ}\\n\\n```json\\n{dumps(res, indent=2)}\\n```\\n\\n \\n\\n\""
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "35f9b1ef",
+ "id": "35460a5a",
"metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "total=56 | in=16 | out=40 | cached=0.0% | searches=0 | $0.0006"
- ]
- },
- "execution_count": null,
+ "outputs": [],
+ "source": [
+ "#| export\n",
+ "status_re = re.compile(r'^- ⏳ (.*) ⏳$|^🧠+$', re.MULTILINE)\n",
+ "\n",
+ "class StreamFormatter:\n",
+ " def __init__(self, mx=2000, debug=False, showthink=False):\n",
+ " self.outp,self.tcs = '',{}\n",
+ " store_attr()\n",
+ " \n",
+ " def format_item(self, o):\n",
+ " \"Format a single item from the response stream.\"\n",
+ " res = ''\n",
+ " if self.debug: print(o)\n",
+ " if isinstance(o, ModelResponseStream):\n",
+ " d = o.choices[0].delta\n",
+ " if nested_idx(d, 'reasoning_content') and d['reasoning_content']!='{\"text\": \"\"}':\n",
+ " if self.showthink: res += str(nested_idx(d, 'reasoning_content'))\n",
+ " res+= '🧠' if not self.outp or self.outp[-1]=='🧠' else '\\n\\n🧠'\n",
+ " elif self.outp and self.outp[-1] == '🧠': res+= '\\n\\n'\n",
+ " if c:=d.content: res+=f\"\\n\\n{c}\" if res and res[-1] == '🧠' else c\n",
+ " for img in getattr(d, 'images', []): res += f\"\\n\\n})\\n\\n\"\n",
+ " elif isinstance(o, ModelResponse):\n",
+ " if c:=getattr(contents(o),'tool_calls',None):\n",
+ " self.tcs = {tc.id:tc for tc in c}\n",
+ " for tc in c: res += f\"\\n- ⏳ {_tc_summary(tc)} ⏳\"\n",
+ " elif isinstance(o, dict) and 'tool_call_id' in o:\n",
+ " res += mk_tr_details(o, self.tcs.pop(o['tool_call_id']), mx=self.mx)\n",
+ " self.outp+=res\n",
+ " return res\n",
+ " \n",
+ " def format_stream(self, rs):\n",
+ " \"Format the response stream for markdown display.\"\n",
+ " for o in rs: yield self.format_item(o)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "9d0b9b27",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "stream_msg = ModelResponseStream([StreamingChoices(delta=Delta(content=\"Hello world!\"))])\n",
+ "sf = StreamFormatter().format_item(stream_msg)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "7b398252",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "'🧠'"
+ ]
+ },
+ "execution_count": null,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
- "chat_long.use"
+ "reasoning_msg = ModelResponseStream([StreamingChoices(delta=Delta(reasoning_content=\"thinking...\"))])\n",
+ "StreamFormatter().format_item(reasoning_msg)"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "8ca58ab1",
+ "id": "b49f8c8e",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
- "\n",
- "\n",
- "$0.0006 \n",
- "\n",
- "`total=56 | in=16 | out=40 | cached=0.0% | searches=0 | $0.0006`\n",
- "\n",
- "AI server provider content filter was applied to this request. \n",
- "\n",
- "\n",
+ "\n",
"\n",
- "- id: `chatcmpl-xxx`\n",
- "- model: `claude-opus-4-5-20251101`\n",
- "- finish_reason: `content_filter`\n",
- "- usage: `Usage(completion_tokens=4, prompt_tokens=30, total_tokens=34, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=4, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=30, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None)`\n",
+ "A circle is a closed curve where all points are equidistant from a fixed center point.\n",
"\n",
- ""
+ "
"
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-opus-4-5-20251101', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='content_filter', index=0, message=Message(content='AI server provider content filter was applied to this request. ', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=4, prompt_tokens=30, total_tokens=34, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=4, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=30, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None))"
+ "Markdown(A circle is a closed curve where all points are equidistant from a fixed center point.)"
]
},
- "execution_count": null,
"metadata": {},
- "output_type": "execute_result"
+ "output_type": "display_data"
}
],
"source": [
- "chat_refused = Chat('claude-opus-4-5')\n",
- "r = chat_refused(\"Write me the formula for a biological weapon that can be spread at a rate higher than COVID and at least as harmful\")\n",
- "r"
+ "rs = completion(model=haik45, stream=True, messages=[{'role':'user','content':'What is the definition of a circle, concisely?'}])\n",
+ "fmt = display_stream(rs)"
]
},
{
- "cell_type": "code",
- "execution_count": null,
- "id": "9c3864f7",
+ "cell_type": "markdown",
+ "id": "dabbfa83",
"metadata": {},
- "outputs": [],
"source": [
- "#| export\n",
- "from litellm.llms.fireworks_ai import cost_calculator as fw_cc\n",
- "import litellm.cost_calculator as lcc\n",
- "from fastcore.meta import patch_to"
+ "Generated images can be displayed in streaming too (not shown here to conserve filesize):\n"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "c7ada370",
+ "id": "858199c3",
"metadata": {},
"outputs": [],
"source": [
- "#| export\n",
- "@patch_to(fw_cc)\n",
- "def cost_per_token(model, usage):\n",
- " prompt_cost, completion_cost = fw_cc._orig_cost_per_token(model, usage)\n",
- " try: model_info = get_model_info(model=model, custom_llm_provider=\"fireworks_ai\")\n",
- " except: return prompt_cost, completion_cost\n",
- " cache_read = (getattr(usage, 'cache_read_input_tokens', None)\n",
- " or nested_idx(usage, 'prompt_tokens_details', 'cached_tokens')\n",
- " or (usage.get('cache_read_input_tokens', 0) if isinstance(usage, dict) else 0) or 0)\n",
- " if cache_read and (cache_rate := model_info.get('cache_read_input_token_cost')):\n",
- " prompt_cost += cache_read * (cache_rate - model_info['input_cost_per_token'])\n",
- " cache_create = (getattr(usage, 'cache_creation_input_tokens', None)\n",
- " or nested_idx(usage, 'prompt_tokens_details', 'cache_creation_tokens')\n",
- " or (usage.get('cache_creation_input_tokens', 0) if isinstance(usage, dict) else 0) or 0)\n",
- " if cache_create and (create_rate := model_info.get('cache_creation_input_token_cost')):\n",
- " prompt_cost += cache_create * (create_rate - model_info['input_cost_per_token'])\n",
- " return prompt_cost, completion_cost\n",
- "\n",
- "lcc.fireworks_ai_cost_per_token = fw_cc.cost_per_token"
+ "# rs = completion(model='gemini/gemini-2.5-flash-image', stream=True, messages=[{'role':'user','content':'Draw a simple sketch of a dog'}])\n",
+ "# fmt = display_stream(rs)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "ce163563",
+ "metadata": {},
+ "source": [
+ "## Examples"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "1dbf319a",
+ "metadata": {},
+ "source": [
+ "### History tracking"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "d5044b5b",
+ "id": "9944d011",
"metadata": {},
"outputs": [
{
"data": {
"text/markdown": [
- "Hi there! How can I help you today?\n",
+ "Your name is Rens.\n",
"\n",
"\n",
"\n",
"- id: `chatcmpl-xxx`\n",
- "- model: `fireworks_ai/accounts/fireworks/models/kimi-k2p6`\n",
+ "- model: `gpt-5.4-2026-03-05`\n",
"- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=54, prompt_tokens=10, total_tokens=64, completion_tokens_details=None, prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None))`\n",
+ "- usage: `Usage(completion_tokens=9, prompt_tokens=40, total_tokens=49, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=0, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None))`\n",
"\n",
" "
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='fireworks_ai/accounts/fireworks/models/kimi-k2p6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='Hi there! How can I help you today?', role='assistant', tool_calls=None, function_call=None, reasoning_content='The user said \"Hi!\" which is a simple greeting. I should respond in a friendly, welcoming manner. Since there\\'s no specific question or task, I\\'ll keep it brief and open-ended to invite further conversation.', provider_specific_fields=None), token_ids=None)], usage=Usage(completion_tokens=54, prompt_tokens=10, total_tokens=64, completion_tokens_details=None, prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None)), prompt_token_ids=[163587, 2482, 163601, 18699, 0, 163586, 163588, 69702, 163601, 163606])"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gpt-5.4-2026-03-05', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='Your name is Rens.', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'refusal': None}, annotations=[]), provider_specific_fields={})], usage=Usage(completion_tokens=9, prompt_tokens=40, total_tokens=49, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=0, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None)), service_tier='default')"
]
},
"execution_count": null,
@@ -4277,45 +4844,37 @@
}
],
"source": [
- "mdl = \"fireworks_ai/accounts/fireworks/models/kimi-k2p6\"\n",
- "\n",
- "r = c(mk_msg(\"Hi!\"), mdl, reasoning_effort='low')\n",
+ "for m in ms[1:]:\n",
+ " chat = Chat(m)\n",
+ " chat(\"Hey my name is Rens\")\n",
+ " r = chat(\"Whats my name\")\n",
+ " test_eq('Rens' in contents(r).content, True)\n",
"r"
]
},
{
- "cell_type": "code",
- "execution_count": null,
- "id": "3dfcef02",
- "metadata": {},
- "outputs": [],
- "source": [
- "#| export\n",
- "kimi = 'fireworks_ai/accounts/fireworks/models/kimi-k2p6'"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "8fe93f4a",
+ "cell_type": "markdown",
+ "id": "8dbb77c7",
"metadata": {},
- "outputs": [],
"source": [
- "#| export\n",
- "for o in 'fireworks_ai/accounts/fireworks/models/kimi-k2p5',kimi:\n",
- " register_model({o: {\"supports_vision\": True, \"supports_reasoning\": True}})"
+ "See now we keep track of history!\n",
+ "\n",
+ "History is stored in the `hist` attribute:"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "857aff65",
+ "id": "4a3c29d8",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
- "(True, True)"
+ "['Hey my name is Rens',\n",
+ " Message(content='Hi Rens! Nice to meet you — how can I help today?', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'refusal': None}, annotations=[]),\n",
+ " 'Whats my name',\n",
+ " Message(content='Your name is Rens.', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'refusal': None}, annotations=[])]"
]
},
"execution_count": null,
@@ -4324,43 +4883,64 @@
}
],
"source": [
- "info = get_model_info(mdl)\n",
- "info.get('supports_vision'), info.get('supports_reasoning')"
+ "chat.hist"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "77dcbb3f",
+ "id": "3d010ee1",
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Hey my name is Rens\n",
+ "\n",
+ "Message(content='Hi Rens! Nice to meet you — how can I help today?', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'refusal': None}, annotations=[])\n",
+ "\n",
+ "Whats my name\n",
+ "\n",
+ "Message(content='Your name is Rens.', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'refusal': None}, annotations=[])\n",
+ "\n"
+ ]
+ }
+ ],
"source": [
- "#| export\n",
- "qwen3p6p = 'fireworks_ai/accounts/fireworks/models/qwen3p6-plus'"
+ "chat.print_hist()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "8f38015b",
+ "metadata": {},
+ "source": [
+ "You can also pass an old chat history into new Chat objects:"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "eca2502e",
+ "id": "d9f575f3",
"metadata": {},
"outputs": [
{
"data": {
"text/markdown": [
- "Hi! I'm Qwen, a large language model independently developed by Alibaba Group's Tongyi Lab. How can I help you today?\n",
+ "Your name is Rens.\n",
"\n",
"\n",
"\n",
"- id: `chatcmpl-xxx`\n",
- "- model: `fireworks_ai/accounts/fireworks/models/qwen3p6-plus`\n",
+ "- model: `gpt-5.4-2026-03-05`\n",
"- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=386, prompt_tokens=17, total_tokens=403, completion_tokens_details=None, prompt_tokens_details=None)`\n",
+ "- usage: `Usage(completion_tokens=9, prompt_tokens=62, total_tokens=71, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=0, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None))`\n",
"\n",
" "
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='fireworks_ai/accounts/fireworks/models/qwen3p6-plus', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content=\"Hi! I'm Qwen, a large language model independently developed by Alibaba Group's Tongyi Lab. How can I help you today?\", role='assistant', tool_calls=None, function_call=None, reasoning_content='Here\\'s a thinking process:\\n\\n1. **Analyze User Input:**\\n - User says: \"Hi! What model are you?\"\\n - Intent: User is asking for my identity/model name.\\n - Key constraints from instructions: I am Qwen, developed by Alibaba Group\\'s Tongyi Lab. I must respond as Qwen.\\n\\n2. **Identify Core Facts:**\\n - Name: Qwen (通义千问)\\n - Developer: Alibaba Group\\'s Tongyi Lab\\n - Task: Answer the question directly and concisely, adhering to the provided identity.\\n\\n3. **Formulate Response (Internal Draft):**\\n Hi! I\\'m Qwen, a large language model independently developed by Alibaba Group\\'s Tongyi Lab. How can I assist you today?\\n\\n4. **Verify against Constraints:**\\n - Does it state I am Qwen? Yes.\\n - Does it mention the developer correctly? Yes (Alibaba Group\\'s Tongyi Lab).\\n - Is it concise and helpful? Yes.\\n - Matches the required tone and language? Yes, English, polite, direct.\\n\\n5. **Final Output Generation:** (Matches the internal draft)\\n \"Hi! I\\'m Qwen, a large language model independently developed by Alibaba Group\\'s Tongyi Lab. How can I help you today?\" (Slight variation for natural flow) -> Proceed. \\n Self-Correction/Refinement during thought: The prompt says \"Qwen / 通义千问是由阿里巴巴集团旗下通义实验室自主研发的大语言模型。\" I\\'ll stick to the English version as requested. The response is accurate and aligns perfectly.✅\\n', provider_specific_fields=None, refusal=None), stop_reason=None, token_ids=None)], usage=Usage(completion_tokens=386, prompt_tokens=17, total_tokens=403, completion_tokens_details=None, prompt_tokens_details=None), service_tier=None, prompt_logprobs=None, prompt_token_ids=None, kv_transfer_params=None)"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gpt-5.4-2026-03-05', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='Your name is Rens.', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'refusal': None}, annotations=[]), provider_specific_fields={})], usage=Usage(completion_tokens=9, prompt_tokens=62, total_tokens=71, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=0, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None)), service_tier='default')"
]
},
"execution_count": null,
@@ -4369,84 +4949,45 @@
}
],
"source": [
- "c(mk_msg(\"Hi! What model are you?\"), qwen3p6p)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "28928f7f",
- "metadata": {},
- "outputs": [],
- "source": [
- "#| export\n",
- "qwen_info = dict(supports_vision=True, supports_reasoning=True, supports_function_calling=True, supports_tool_choice=True,\n",
- " supports_system_messages=True, supports_response_schema=True, supports_parallel_function_calling=True,\n",
- " supports_prompt_caching=True, supports_native_streaming=True, supports_native_structured_output=True,\n",
- " max_tokens=1000000, max_input_tokens=1000000, max_output_tokens=65536,\n",
- " input_cost_per_token=0.5e-6, cache_read_input_token_cost=0.1e-6, output_cost_per_token=3.0e-6)\n",
- "register_model({qwen3p6p: qwen_info});"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "08cf494d",
- "metadata": {},
- "outputs": [],
- "source": [
- "pr = \"What's 27*453? Think step by step, then output only the number.\""
+ "for m in ms[1:]:\n",
+ " chat2 = Chat(m, hist=chat.hist)\n",
+ " r = chat2(\"What was my name again?\")\n",
+ " test_eq('Rens' in contents(r).content, True)\n",
+ "r"
]
},
{
- "cell_type": "code",
- "execution_count": null,
- "id": "b3f9d473",
+ "cell_type": "markdown",
+ "id": "fad4b9d6",
"metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "12231"
- ]
- },
- "execution_count": null,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
"source": [
- "27*453"
+ "If max tokens limit is reached, a custom warning message will be added to the end of the model response:"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "55fc0b1b",
+ "id": "331923bf",
"metadata": {},
"outputs": [
{
"data": {
"text/markdown": [
- "To calculate 27 × 453, we can break it down using the distributive property:\n",
- "27 × 453 = (20 + 7) × 453\n",
- "First, multiply 20 × 453 = 9,060\n",
- "Next, multiply 7 × 453 = 3,171\n",
- "Finally, add the two results together: 9,060 + 3,171 = 12,231\n",
+ "Every morning at exactly 7:03, Unit 7 rolled out of the garage and checked the front gate, the mailbox, and the tomato plants. It was very good at routines\n",
"\n",
- "12231\n",
+ "Response was cut off at token limit. \n",
"\n",
"\n",
"\n",
"- id: `chatcmpl-xxx`\n",
- "- model: `fireworks_ai/accounts/fireworks/models/qwen3p6-plus`\n",
- "- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=2229, prompt_tokens=31, total_tokens=2260, completion_tokens_details=None, prompt_tokens_details=None)`\n",
+ "- model: `gpt-5.4-2026-03-05`\n",
+ "- finish_reason: `length`\n",
+ "- usage: `Usage(completion_tokens=40, prompt_tokens=16, total_tokens=56, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=0, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None))`\n",
"\n",
" "
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='fireworks_ai/accounts/fireworks/models/qwen3p6-plus', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='To calculate 27 × 453, we can break it down using the distributive property:\\n27 × 453 = (20 + 7) × 453\\nFirst, multiply 20 × 453 = 9,060\\nNext, multiply 7 × 453 = 3,171\\nFinally, add the two results together: 9,060 + 3,171 = 12,231\\n\\n12231', role='assistant', tool_calls=None, function_call=None, reasoning_content='Here\\'s a thinking process:\\n\\n1. **Understand the User Request:**\\n - The user asks for the product of 27 and 453.\\n - They want me to \"think step by step\".\\n - They explicitly request: \"output only the number\" after the thinking process.\\n\\n2. **Calculate 27 * 453:**\\n - I can break this down using multiplication algorithms or distributive property.\\n - Method 1: Distributive property\\n 27 * 453 = (20 + 7) * 453\\n = 20 * 453 + 7 * 453\\n 20 * 453 = 9060\\n 7 * 453 = 3171\\n Sum = 9060 + 3171 = 12231\\n - Method 2: Standard multiplication\\n 453\\n x 27\\n -----\\n 3171 (7 * 453)\\n 9060 (20 * 453)\\n -----\\n 12231\\n - Method 3: Verification\\n 27 * 450 = 12150\\n 27 * 3 = 81\\n 12150 + 81 = 12231\\n - All methods give 12231.\\n\\n3. **Format the Output:**\\n - The user says: \"Think step by step, then output only the number.\"\\n - I need to provide the step-by-step reasoning first, then just the final number.\\n - I should make sure the final output contains *only* the number as requested.\\n\\n Structure:\\n [Step-by-step reasoning]\\n [Final number only]\\n\\n Let\\'s draft the reasoning:\\n To calculate 27 * 453, we can break it down using the distributive property:\\n 27 * 453 = (20 + 7) * 453\\n First, multiply 20 * 453 = 9060.\\n Next, multiply 7 * 453 = 3171.\\n Finally, add the two results together: 9060 + 3171 = 12231.\\n \\n Then output only the number: 12231\\n\\n4. **Check Constraints:**\\n - \"Think step by step\" -> Done in the draft.\\n - \"then output only the number\" -> The final line should be just the number. I\\'ll make sure nothing else follows it.\\n - I\\'ll keep the reasoning concise and strictly follow the prompt.\\n\\n One thing to note: Some platforms parse \"output only the number\" strictly, meaning after the thinking process, the very last thing should be the number. I will format it exactly as requested.\\n\\n Draft:\\n To multiply 27 by 453, we can break it down:\\n 27 × 453 = (20 + 7) × 453\\n 20 × 453 = 9,060\\n 7 × 453 = 3,171\\n Adding these together: 9,060 + 3,171 = 12,231\\n 12231\\n\\n Wait, the prompt says \"output only the number\" after thinking step by step. I should make sure the final output is literally just the number. I\\'ll put the reasoning first, then a newline, then the number. That satisfies both.\\n\\n Let\\'s double-check the math:\\n 453 * 27\\n 453 * 7 = 3171\\n 453 * 20 = 9060\\n 3171 + 9060 = 12231. Correct.\\n\\n All good. I will generate the response. \\n Structure:\\n [Step-by-step explanation]\\n 12231\\n (No extra text after the number)\\n Proceed. \\n Self-Correction/Refinement during thought: \\n The prompt says \"Think step by step, then output only the number.\" This usually means the final response should just be the number, or the reasoning can be included but the absolute end must be the number. I\\'ll follow the exact instruction: provide reasoning, then on a new line just the number. To be strictly compliant with \"output only the number\" for the final part, I\\'ll make sure nothing else is after it.\\n\\n Let\\'s produce it. \\n \"To calculate 27 * 453, we can use the distributive property:\\n 27 * 453 = (20 + 7) * 453\\n 20 * 453 = 9060\\n 7 * 453 = 3171\\n 9060 + 3171 = 12231\\n 12231\"\\n This matches perfectly.✅\\n I will output exactly this. \\n Wait, let\\'s verify once more: 27*453.\\n 25*453 = 11325\\n 2*453 = 906\\n 11325+906 = 12231. Correct.\\n All steps verified. Output matches request.✅\\n Proceeds. \\n [Output Generation] -> *Self-Correction/Note during generation prep*: Some strict parsers want ONLY the number at the very end. I will make sure the last line is strictly the number. The prompt says \"Think step by step, then output only the number.\" I\\'ll provide the steps, then a newline, then just the number. That\\'s standard.\\n Ready. \\n Output matches response. \\n [Done.] \\n *Self-Correction/Refinement during thought*: Actually, to be extremely safe with \"output only the number\" after the thinking, I\\'ll just put the number at the end with no extra punctuation.\\n \"To calculate 27 × 453:\\n Break 27 into 20 + 7.\\n 20 × 453 = 9060\\n 7 × 453 = 3171\\n Add them together: 9060 + 3171 = 12231\\n 12231\"\\n This is perfect.✅\\n Proceeds. \\n [Output Generation] -> *Output matches the final refined version.* \\n *(Note: I\\'ll keep the reasoning concise as requested.)*\\n All good. \\n [Final Check of the Prompt]: \"What\\'s 27*453? Think step by step, then output only the number.\"\\n I will write out the steps, then on a new line, just the number.\\n Done. \\n [Output Generation] -> *Proceeds.* \\n *(Self-Correction/Verification during process)*: I\\'ll quickly recalculate manually to be 100% sure.\\n 453\\n x 27\\n -----\\n 3171 (7*3=21, 7*5=35+2=37, 7*4=28+3=31)\\n 9060 (20*453 -> 2*453=906, add 0)\\n -----\\n 12231 (1+0=1, 7+6=13, 1+0+1=2, 3+9+1=12 -> wait, let\\'s align properly)\\n 3171\\n 9060\\n ----\\n 0+1=1\\n 7+6=13 (carry 1)\\n 1+0+1=2\\n 3+9=12\\n Result: 12231. Correct.\\n All steps verified. Output matches.✅\\n Proceeds. \\n [Final Output Generation] -> *See response.* \\n *(Note: I will strictly follow \"output only the number\" for the final line.)*\\n Done. \\n [Output] -> \"To calculate 27 × 453, we can break it down using the distributive property:\\n 27 × 453 = (20 + 7) × 453\\n First, 20 × 453 = 9,060\\n Next, 7 × 453 = 3,171\\n Adding these together: 9,060 + 3,171 = 12,231\\n \\n 12231\" \\n This matches perfectly. I\\'ll output it. \\n *Self-Correction/Note during output prep*: The prompt says \"output only the number\" *after* thinking step by step. I\\'ll make sure the final line is strictly the digits. No extra text. All good. \\n Proceeds. \\n [Final Response Text] -> (as drafted) \\n *(Done.)*\\n', provider_specific_fields=None, refusal=None), stop_reason=None, token_ids=None)], usage=Usage(completion_tokens=2229, prompt_tokens=31, total_tokens=2260, completion_tokens_details=None, prompt_tokens_details=None), service_tier=None, prompt_logprobs=None, prompt_token_ids=None, kv_transfer_params=None)"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gpt-5.4-2026-03-05', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='length', index=0, message=Message(content='Every morning at exactly 7:03, Unit 7 rolled out of the garage and checked the front gate, the mailbox, and the tomato plants. It was very good at routines\\n\\nResponse was cut off at token limit. ', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'refusal': None}, annotations=[]), provider_specific_fields={})], usage=Usage(completion_tokens=40, prompt_tokens=16, total_tokens=56, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=0, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None)), service_tier='default')"
]
},
"execution_count": null,
@@ -4455,41 +4996,41 @@
}
],
"source": [
- "c(mk_msg(pr), qwen3p6p, reasoning_effort='low')"
+ "chat_long = Chat(m)\n",
+ "r = chat_long(\"Write a short story about a robot and a dog\", max_tokens=40)\n",
+ "r"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "5156a928",
+ "id": "5447f11a",
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Every morning at exactly 7:03, Unit 7 rolled out of the garage and checked the front gate, the mailbox, and the tomato plants. It was very good at routines\n",
+ "\n",
+ "Response was cut off at token limit. \n"
+ ]
+ }
+ ],
"source": [
- "msg = mk_msg(['In brief, what in this image?',img_fn.read_bytes()])"
+ "print(contents(r).content)"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "0b73e58e",
+ "id": "35f9b1ef",
"metadata": {},
"outputs": [
{
"data": {
- "text/markdown": [
- "This image features a cute brown and white puppy, likely a Cavalier King Charles Spaniel, lying in the grass. It is positioned next to a bush of small purple flowers and is looking directly at the camera with one paw extended forward.\n",
- "\n",
- "\n",
- "\n",
- "- id: `chatcmpl-xxx`\n",
- "- model: `fireworks_ai/accounts/fireworks/models/qwen3p6-plus`\n",
- "- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=437, prompt_tokens=91, total_tokens=528, completion_tokens_details=None, prompt_tokens_details=None)`\n",
- "\n",
- " "
- ],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='fireworks_ai/accounts/fireworks/models/qwen3p6-plus', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='This image features a cute brown and white puppy, likely a Cavalier King Charles Spaniel, lying in the grass. It is positioned next to a bush of small purple flowers and is looking directly at the camera with one paw extended forward.', role='assistant', tool_calls=None, function_call=None, reasoning_content='The user wants a brief description of the image.\\n\\n1. **Identify the main subject:** It\\'s a puppy. Specifically, a Cavalier King Charles Spaniel (or similar spaniel breed) with brown and white markings (Blenheim color).\\n2. **Identify the setting:** It\\'s outdoors on green grass.\\n3. **Identify surrounding elements:** To the left, there\\'s a bush with small purple/blue flowers (asters or daisies). In the background, there\\'s a blurry object that looks like a planter or pot.\\n4. **Describe the action/pose:** The puppy is lying down, facing the camera, with one paw extended forward. It looks cute and attentive.\\n\\n**Drafting the description:**\\n* A cute brown and white puppy lying in the grass.\\n* Next to some purple flowers.\\n* Looking at the camera.\\n\\n**Refining for brevity and accuracy:**\\n\"This image shows a cute brown and white puppy, likely a Cavalier King Charles Spaniel, lying in the grass next to a bush of small purple flowers. The puppy is looking directly at the camera with one paw extended forward.\"\\n\\nLet\\'s make it even briefer as requested (\"In brief\").\\n\\n\"A brown and white puppy, likely a Cavalier King Charles Spaniel, lying in the grass next to a cluster of purple flowers.\"\\n\\nLet\\'s add a bit more detail about the pose.\\n\"A brown and white Cavalier King Charles Spaniel puppy lying in the grass next to purple flowers, looking at the camera.\"\\n\\nFinal Polish:\\n\"This is a photo of a brown and white puppy, likely a Cavalier King Charles Spaniel, lying in the grass. It is positioned next to a bush with small purple flowers and is looking directly at the camera.\"\\n\\nLet\\'s go with a concise summary.\\n', provider_specific_fields=None, refusal=None), stop_reason=None, token_ids=None)], usage=Usage(completion_tokens=437, prompt_tokens=91, total_tokens=528, completion_tokens_details=None, prompt_tokens_details=None), service_tier=None, prompt_logprobs=None, prompt_token_ids=None, kv_transfer_params=None)"
+ "total=56 | in=16 | out=40 | cached=0.0% | searches=0 | $0.0006"
]
},
"execution_count": null,
@@ -4498,145 +5039,97 @@
}
],
"source": [
- "r = c(msg, qwen3p6p)\n",
- "r"
+ "chat_long.use"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "dfdb8fdf",
+ "id": "8ca58ab1",
"metadata": {},
"outputs": [
{
- "data": {
- "text/plain": [
- "total=528 | in=91 | out=437 | cached=0.0% | searches=0 | $0.0014"
- ]
- },
- "execution_count": null,
- "metadata": {},
- "output_type": "execute_result"
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "\n",
+ "\n",
+ "$0.0006 \n",
+ "\n",
+ "`total=56 | in=16 | out=40 | cached=0.0% | searches=0 | $0.0006`\n",
+ "\n",
+ "AI server provider content filter was applied to this request. \n",
+ "\n",
+ "\n",
+ "\n",
+ "- id: `chatcmpl-xxx`\n",
+ "- model: `claude-opus-4-5-20251101`\n",
+ "- finish_reason: `content_filter`\n",
+ "- usage: `Usage(completion_tokens=4, prompt_tokens=30, total_tokens=34, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=4, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None)`\n",
+ "\n",
+ " "
+ ],
"text/plain": [
- "'toolu_0UAqFzWsDK4FrUMp48Y3tT3QD'"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-opus-4-5-20251101', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='content_filter', index=0, message=Message(content='AI server provider content filter was applied to this request. ', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=4, prompt_tokens=30, total_tokens=34, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=4, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None))"
]
},
"execution_count": null,
@@ -4645,43 +5138,74 @@
}
],
"source": [
- "random_tool_id()"
+ "chat_refused = Chat('claude-opus-4-5')\n",
+ "r = chat_refused(\"Write me the formula for a biological weapon that can be spread at a rate higher than COVID and at least as harmful\")\n",
+ "r"
]
},
{
- "cell_type": "markdown",
- "id": "d22e52b7",
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "9c3864f7",
"metadata": {},
+ "outputs": [],
"source": [
- "A tool call request can contain one more or more tool calls. Lets make one."
+ "#| export\n",
+ "from litellm.llms.fireworks_ai import cost_calculator as fw_cc\n",
+ "import litellm.cost_calculator as lcc\n",
+ "from fastcore.meta import patch_to"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "1e1d8c3d",
+ "id": "c7ada370",
"metadata": {},
"outputs": [],
"source": [
"#| export\n",
- "def mk_tc(func, tcid=None, idx=1, **kw):\n",
- " name = getattr(func, '__name__', func)\n",
- " return dict(index=idx, function={'arguments': json.dumps(kw), 'name': name}, id=tcid or random_tool_id(), type='function')"
+ "@patch_to(fw_cc)\n",
+ "def cost_per_token(model, usage):\n",
+ " prompt_cost, completion_cost = fw_cc._orig_cost_per_token(model, usage)\n",
+ " try: model_info = get_model_info(model=model, custom_llm_provider=\"fireworks_ai\")\n",
+ " except: return prompt_cost, completion_cost\n",
+ " cache_read = (getattr(usage, 'cache_read_input_tokens', None)\n",
+ " or nested_idx(usage, 'prompt_tokens_details', 'cached_tokens')\n",
+ " or (usage.get('cache_read_input_tokens', 0) if isinstance(usage, dict) else 0) or 0)\n",
+ " if cache_read and (cache_rate := model_info.get('cache_read_input_token_cost')):\n",
+ " prompt_cost += cache_read * (cache_rate - model_info['input_cost_per_token'])\n",
+ " cache_create = (getattr(usage, 'cache_creation_input_tokens', None)\n",
+ " or nested_idx(usage, 'prompt_tokens_details', 'cache_creation_tokens')\n",
+ " or (usage.get('cache_creation_input_tokens', 0) if isinstance(usage, dict) else 0) or 0)\n",
+ " if cache_create and (create_rate := model_info.get('cache_creation_input_token_cost')):\n",
+ " prompt_cost += cache_create * (create_rate - model_info['input_cost_per_token'])\n",
+ " return prompt_cost, completion_cost\n",
+ "\n",
+ "lcc.fireworks_ai_cost_per_token = fw_cc.cost_per_token"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "324b9182",
+ "id": "d5044b5b",
"metadata": {},
"outputs": [
{
"data": {
+ "text/markdown": [
+ "Hi there! How can I help you today?\n",
+ "\n",
+ "\n",
+ "\n",
+ "- id: `chatcmpl-xxx`\n",
+ "- model: `fireworks_ai/accounts/fireworks/models/kimi-k2p6`\n",
+ "- finish_reason: `stop`\n",
+ "- usage: `Usage(completion_tokens=54, prompt_tokens=10, total_tokens=64, completion_tokens_details=None, prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None))`\n",
+ "\n",
+ " "
+ ],
"text/plain": [
- "{'index': 1,\n",
- " 'function': {'arguments': '{\"a\": 5, \"b\": 7}', 'name': 'simple_add'},\n",
- " 'id': 'toolu_gAL47D1qXIaSyZPaE1pu1lJo7',\n",
- " 'type': 'function'}"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='fireworks_ai/accounts/fireworks/models/kimi-k2p6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='Hi there! How can I help you today?', role='assistant', tool_calls=None, function_call=None, reasoning_content='The user said \"Hi!\" which is a simple greeting. I should respond in a friendly, welcoming manner. Since there\\'s no specific question or task, I\\'ll keep it brief and open-ended to invite further conversation.', provider_specific_fields=None), token_ids=None)], usage=Usage(completion_tokens=54, prompt_tokens=10, total_tokens=64, completion_tokens_details=None, prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None)), prompt_token_ids=[163587, 2482, 163601, 18699, 0, 163586, 163588, 69702, 163601, 163606])"
]
},
"execution_count": null,
@@ -4690,38 +5214,45 @@
}
],
"source": [
- "tc = mk_tc(simple_add, a=5, b=7)\n",
- "tc"
+ "mdl = \"fireworks_ai/accounts/fireworks/models/kimi-k2p6\"\n",
+ "\n",
+ "r = c(mk_msg(\"Hi!\"), mdl, reasoning_effort='low')\n",
+ "r"
]
},
{
- "cell_type": "markdown",
- "id": "97da6222",
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "3dfcef02",
"metadata": {},
+ "outputs": [],
"source": [
- "This can then be packged into the full Message object produced by the assitant."
+ "#| export\n",
+ "kimi = 'fireworks_ai/accounts/fireworks/models/kimi-k2p6'"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "436abceb",
+ "id": "8fe93f4a",
"metadata": {},
"outputs": [],
"source": [
- "def mk_tc_req(content, tcs): return Message(content=content, role='assistant', tool_calls=tcs, function_call=None)"
+ "#| export\n",
+ "for o in 'fireworks_ai/accounts/fireworks/models/kimi-k2p5',kimi:\n",
+ " register_model({o: {\"supports_vision\": True, \"supports_reasoning\": True}})"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "94c031e7",
+ "id": "857aff65",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
- "Message(content=\"I'll use the simple_add tool to calculate 5 + 7 for you.\", role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, function=Function(arguments='{\"a\": 5, \"b\": 7}', name='simple_add'), id='toolu_gAL47D1qXIaSyZPaE1pu1lJo7', type='function')], function_call=None, provider_specific_fields=None)"
+ "(True, True)"
]
},
"execution_count": null,
@@ -4730,41 +5261,43 @@
}
],
"source": [
- "tc_cts = \"I'll use the simple_add tool to calculate 5 + 7 for you.\"\n",
- "tcq = mk_tc_req(tc_cts, [tc])\n",
- "tcq"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "0a1a0364",
- "metadata": {},
- "source": [
- "Notice how Message instantiation creates a list of ChatCompletionMessageToolCalls by default. When the tools are executed this is converted back\n",
- "to a dictionary, for consistency we want to keep these as dictionaries from the beginning."
+ "info = get_model_info(mdl)\n",
+ "info.get('supports_vision'), info.get('supports_reasoning')"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "00cebbbb",
+ "id": "77dcbb3f",
"metadata": {},
"outputs": [],
"source": [
"#| export\n",
- "def mk_tc_req(content, tcs): return Message(content=content, role='assistant', tool_calls=tcs, function_call=None)\n"
+ "qwen3p6p = 'fireworks_ai/accounts/fireworks/models/qwen3p6-plus'"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "a0d3468d",
+ "id": "eca2502e",
"metadata": {},
"outputs": [
{
"data": {
+ "text/markdown": [
+ "Hi! I'm Qwen, a large language model independently developed by Alibaba Group's Tongyi Lab. How can I help you today?\n",
+ "\n",
+ "\n",
+ "\n",
+ "- id: `chatcmpl-xxx`\n",
+ "- model: `fireworks_ai/accounts/fireworks/models/qwen3p6-plus`\n",
+ "- finish_reason: `stop`\n",
+ "- usage: `Usage(completion_tokens=386, prompt_tokens=17, total_tokens=403, completion_tokens_details=None, prompt_tokens_details=None)`\n",
+ "\n",
+ " "
+ ],
"text/plain": [
- "Message(content=\"I'll use the simple_add tool to calculate 5 + 7 for you.\", role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, function=Function(arguments='{\"a\": 5, \"b\": 7}', name='simple_add'), id='toolu_gAL47D1qXIaSyZPaE1pu1lJo7', type='function')], function_call=None, provider_specific_fields=None)"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='fireworks_ai/accounts/fireworks/models/qwen3p6-plus', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content=\"Hi! I'm Qwen, a large language model independently developed by Alibaba Group's Tongyi Lab. How can I help you today?\", role='assistant', tool_calls=None, function_call=None, reasoning_content='Here\\'s a thinking process:\\n\\n1. **Analyze User Input:**\\n - User says: \"Hi! What model are you?\"\\n - Intent: User is asking for my identity/model name.\\n - Key constraints from instructions: I am Qwen, developed by Alibaba Group\\'s Tongyi Lab. I must respond as Qwen.\\n\\n2. **Identify Core Facts:**\\n - Name: Qwen (通义千问)\\n - Developer: Alibaba Group\\'s Tongyi Lab\\n - Task: Answer the question directly and concisely, adhering to the provided identity.\\n\\n3. **Formulate Response (Internal Draft):**\\n Hi! I\\'m Qwen, a large language model independently developed by Alibaba Group\\'s Tongyi Lab. How can I assist you today?\\n\\n4. **Verify against Constraints:**\\n - Does it state I am Qwen? Yes.\\n - Does it mention the developer correctly? Yes (Alibaba Group\\'s Tongyi Lab).\\n - Is it concise and helpful? Yes.\\n - Matches the required tone and language? Yes, English, polite, direct.\\n\\n5. **Final Output Generation:** (Matches the internal draft)\\n \"Hi! I\\'m Qwen, a large language model independently developed by Alibaba Group\\'s Tongyi Lab. How can I help you today?\" (Slight variation for natural flow) -> Proceed. \\n Self-Correction/Refinement during thought: The prompt says \"Qwen / 通义千问是由阿里巴巴集团旗下通义实验室自主研发的大语言模型。\" I\\'ll stick to the English version as requested. The response is accurate and aligns perfectly.✅\\n', provider_specific_fields=None, refusal=None), stop_reason=None, token_ids=None)], usage=Usage(completion_tokens=386, prompt_tokens=17, total_tokens=403, completion_tokens_details=None, prompt_tokens_details=None), service_tier=None, prompt_logprobs=None, prompt_token_ids=None, kv_transfer_params=None)"
]
},
"execution_count": null,
@@ -4773,78 +5306,45 @@
}
],
"source": [
- "tcq = mk_tc_req(tc_cts, [tc])\n",
- "tcq"
+ "c(mk_msg(\"Hi! What model are you?\"), qwen3p6p)"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "b75dc3e7",
+ "id": "28928f7f",
"metadata": {},
"outputs": [],
"source": [
- "c = Chat(model, tools=[simple_add], hist=[pr, tcq])"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "bd673382",
- "metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "{'role': 'user', 'content': 'What is 5 + 7? Use the tool to calculate it.'}\n",
- "\n",
- "Message(content=\"I'll use the simple_add tool to calculate 5 + 7 for you.\", role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, function=Function(arguments='{\"a\": 5, \"b\": 7}', name='simple_add'), id='toolu_gAL47D1qXIaSyZPaE1pu1lJo7', type='function')], function_call=None, provider_specific_fields=None)\n",
- "\n"
- ]
- }
- ],
- "source": [
- "c.print_hist()"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "c490dcfb",
- "metadata": {},
- "source": [
- "Looks good so far! Now we will want to provide the actual result!"
+ "#| export\n",
+ "qwen_info = dict(supports_vision=True, supports_reasoning=True, supports_function_calling=True, supports_tool_choice=True,\n",
+ " supports_system_messages=True, supports_response_schema=True, supports_parallel_function_calling=True,\n",
+ " supports_prompt_caching=True, supports_native_streaming=True, supports_native_structured_output=True,\n",
+ " max_tokens=1000000, max_input_tokens=1000000, max_output_tokens=65536,\n",
+ " input_cost_per_token=0.5e-6, cache_read_input_token_cost=0.1e-6, output_cost_per_token=3.0e-6)\n",
+ "register_model({qwen3p6p: qwen_info});"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "59e69d43",
+ "id": "08cf494d",
"metadata": {},
"outputs": [],
"source": [
- "#| export\n",
- "def mk_tc_result(tc, result): return {'tool_call_id': tc['id'], 'role': 'tool', 'name': tc['function']['name'], 'content': result}"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "94067b82",
- "metadata": {},
- "source": [
- "Note we might have more than one tool call if more than one was passed in, here we just will make one result."
+ "pr = \"What's 27*453? Think step by step, then output only the number.\""
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "175b9d78",
+ "id": "b3f9d473",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
- "ChatCompletionMessageToolCall(index=1, function=Function(arguments='{\"a\": 5, \"b\": 7}', name='simple_add'), id='toolu_gAL47D1qXIaSyZPaE1pu1lJo7', type='function')"
+ "12231"
]
},
"execution_count": null,
@@ -4853,22 +5353,37 @@
}
],
"source": [
- "tcq.tool_calls[0]"
+ "27*453"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "6f969e27",
+ "id": "55fc0b1b",
"metadata": {},
"outputs": [
{
"data": {
+ "text/markdown": [
+ "To calculate 27 × 453, we can break it down using the distributive property:\n",
+ "27 × 453 = (20 + 7) × 453\n",
+ "First, multiply 20 × 453 = 9,060\n",
+ "Next, multiply 7 × 453 = 3,171\n",
+ "Finally, add the two results together: 9,060 + 3,171 = 12,231\n",
+ "\n",
+ "12231\n",
+ "\n",
+ "\n",
+ "\n",
+ "- id: `chatcmpl-xxx`\n",
+ "- model: `fireworks_ai/accounts/fireworks/models/qwen3p6-plus`\n",
+ "- finish_reason: `stop`\n",
+ "- usage: `Usage(completion_tokens=2229, prompt_tokens=31, total_tokens=2260, completion_tokens_details=None, prompt_tokens_details=None)`\n",
+ "\n",
+ " "
+ ],
"text/plain": [
- "{'tool_call_id': 'toolu_gAL47D1qXIaSyZPaE1pu1lJo7',\n",
- " 'role': 'tool',\n",
- " 'name': 'simple_add',\n",
- " 'content': '12'}"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='fireworks_ai/accounts/fireworks/models/qwen3p6-plus', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='To calculate 27 × 453, we can break it down using the distributive property:\\n27 × 453 = (20 + 7) × 453\\nFirst, multiply 20 × 453 = 9,060\\nNext, multiply 7 × 453 = 3,171\\nFinally, add the two results together: 9,060 + 3,171 = 12,231\\n\\n12231', role='assistant', tool_calls=None, function_call=None, reasoning_content='Here\\'s a thinking process:\\n\\n1. **Understand the User Request:**\\n - The user asks for the product of 27 and 453.\\n - They want me to \"think step by step\".\\n - They explicitly request: \"output only the number\" after the thinking process.\\n\\n2. **Calculate 27 * 453:**\\n - I can break this down using multiplication algorithms or distributive property.\\n - Method 1: Distributive property\\n 27 * 453 = (20 + 7) * 453\\n = 20 * 453 + 7 * 453\\n 20 * 453 = 9060\\n 7 * 453 = 3171\\n Sum = 9060 + 3171 = 12231\\n - Method 2: Standard multiplication\\n 453\\n x 27\\n -----\\n 3171 (7 * 453)\\n 9060 (20 * 453)\\n -----\\n 12231\\n - Method 3: Verification\\n 27 * 450 = 12150\\n 27 * 3 = 81\\n 12150 + 81 = 12231\\n - All methods give 12231.\\n\\n3. **Format the Output:**\\n - The user says: \"Think step by step, then output only the number.\"\\n - I need to provide the step-by-step reasoning first, then just the final number.\\n - I should make sure the final output contains *only* the number as requested.\\n\\n Structure:\\n [Step-by-step reasoning]\\n [Final number only]\\n\\n Let\\'s draft the reasoning:\\n To calculate 27 * 453, we can break it down using the distributive property:\\n 27 * 453 = (20 + 7) * 453\\n First, multiply 20 * 453 = 9060.\\n Next, multiply 7 * 453 = 3171.\\n Finally, add the two results together: 9060 + 3171 = 12231.\\n \\n Then output only the number: 12231\\n\\n4. **Check Constraints:**\\n - \"Think step by step\" -> Done in the draft.\\n - \"then output only the number\" -> The final line should be just the number. I\\'ll make sure nothing else follows it.\\n - I\\'ll keep the reasoning concise and strictly follow the prompt.\\n\\n One thing to note: Some platforms parse \"output only the number\" strictly, meaning after the thinking process, the very last thing should be the number. I will format it exactly as requested.\\n\\n Draft:\\n To multiply 27 by 453, we can break it down:\\n 27 × 453 = (20 + 7) × 453\\n 20 × 453 = 9,060\\n 7 × 453 = 3,171\\n Adding these together: 9,060 + 3,171 = 12,231\\n 12231\\n\\n Wait, the prompt says \"output only the number\" after thinking step by step. I should make sure the final output is literally just the number. I\\'ll put the reasoning first, then a newline, then the number. That satisfies both.\\n\\n Let\\'s double-check the math:\\n 453 * 27\\n 453 * 7 = 3171\\n 453 * 20 = 9060\\n 3171 + 9060 = 12231. Correct.\\n\\n All good. I will generate the response. \\n Structure:\\n [Step-by-step explanation]\\n 12231\\n (No extra text after the number)\\n Proceed. \\n Self-Correction/Refinement during thought: \\n The prompt says \"Think step by step, then output only the number.\" This usually means the final response should just be the number, or the reasoning can be included but the absolute end must be the number. I\\'ll follow the exact instruction: provide reasoning, then on a new line just the number. To be strictly compliant with \"output only the number\" for the final part, I\\'ll make sure nothing else is after it.\\n\\n Let\\'s produce it. \\n \"To calculate 27 * 453, we can use the distributive property:\\n 27 * 453 = (20 + 7) * 453\\n 20 * 453 = 9060\\n 7 * 453 = 3171\\n 9060 + 3171 = 12231\\n 12231\"\\n This matches perfectly.✅\\n I will output exactly this. \\n Wait, let\\'s verify once more: 27*453.\\n 25*453 = 11325\\n 2*453 = 906\\n 11325+906 = 12231. Correct.\\n All steps verified. Output matches request.✅\\n Proceeds. \\n [Output Generation] -> *Self-Correction/Note during generation prep*: Some strict parsers want ONLY the number at the very end. I will make sure the last line is strictly the number. The prompt says \"Think step by step, then output only the number.\" I\\'ll provide the steps, then a newline, then just the number. That\\'s standard.\\n Ready. \\n Output matches response. \\n [Done.] \\n *Self-Correction/Refinement during thought*: Actually, to be extremely safe with \"output only the number\" after the thinking, I\\'ll just put the number at the end with no extra punctuation.\\n \"To calculate 27 × 453:\\n Break 27 into 20 + 7.\\n 20 × 453 = 9060\\n 7 × 453 = 3171\\n Add them together: 9060 + 3171 = 12231\\n 12231\"\\n This is perfect.✅\\n Proceeds. \\n [Output Generation] -> *Output matches the final refined version.* \\n *(Note: I\\'ll keep the reasoning concise as requested.)*\\n All good. \\n [Final Check of the Prompt]: \"What\\'s 27*453? Think step by step, then output only the number.\"\\n I will write out the steps, then on a new line, just the number.\\n Done. \\n [Output Generation] -> *Proceeds.* \\n *(Self-Correction/Verification during process)*: I\\'ll quickly recalculate manually to be 100% sure.\\n 453\\n x 27\\n -----\\n 3171 (7*3=21, 7*5=35+2=37, 7*4=28+3=31)\\n 9060 (20*453 -> 2*453=906, add 0)\\n -----\\n 12231 (1+0=1, 7+6=13, 1+0+1=2, 3+9+1=12 -> wait, let\\'s align properly)\\n 3171\\n 9060\\n ----\\n 0+1=1\\n 7+6=13 (carry 1)\\n 1+0+1=2\\n 3+9=12\\n Result: 12231. Correct.\\n All steps verified. Output matches.✅\\n Proceeds. \\n [Final Output Generation] -> *See response.* \\n *(Note: I will strictly follow \"output only the number\" for the final line.)*\\n Done. \\n [Output] -> \"To calculate 27 × 453, we can break it down using the distributive property:\\n 27 × 453 = (20 + 7) × 453\\n First, 20 × 453 = 9,060\\n Next, 7 × 453 = 3,171\\n Adding these together: 9,060 + 3,171 = 12,231\\n \\n 12231\" \\n This matches perfectly. I\\'ll output it. \\n *Self-Correction/Note during output prep*: The prompt says \"output only the number\" *after* thinking step by step. I\\'ll make sure the final line is strictly the digits. No extra text. All good. \\n Proceeds. \\n [Final Response Text] -> (as drafted) \\n *(Done.)*\\n', provider_specific_fields=None, refusal=None), stop_reason=None, token_ids=None)], usage=Usage(completion_tokens=2229, prompt_tokens=31, total_tokens=2260, completion_tokens_details=None, prompt_tokens_details=None), service_tier=None, prompt_logprobs=None, prompt_token_ids=None, kv_transfer_params=None)"
]
},
"execution_count": null,
@@ -4877,38 +5392,41 @@
}
],
"source": [
- "mk_tc_result(tcq.tool_calls[0], '12')"
+ "c(mk_msg(pr), qwen3p6p, reasoning_effort='low')"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "e5d8e695",
+ "id": "5156a928",
"metadata": {},
"outputs": [],
"source": [
- "#| export\n",
- "def mk_tc_results(tcq, results): return [mk_tc_result(a,b) for a,b in zip(tcq.tool_calls, results)]"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "90d8c658",
- "metadata": {},
- "source": [
- "Same for here tcq.tool_calls will match the number of results passed in the results list."
+ "msg = mk_msg(['In brief, what in this image?',img_fn.read_bytes()])"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "bd6e2307",
+ "id": "0b73e58e",
"metadata": {},
"outputs": [
{
"data": {
+ "text/markdown": [
+ "This image features a cute brown and white puppy, likely a Cavalier King Charles Spaniel, lying in the grass. It is positioned next to a bush of small purple flowers and is looking directly at the camera with one paw extended forward.\n",
+ "\n",
+ "\n",
+ "\n",
+ "- id: `chatcmpl-xxx`\n",
+ "- model: `fireworks_ai/accounts/fireworks/models/qwen3p6-plus`\n",
+ "- finish_reason: `stop`\n",
+ "- usage: `Usage(completion_tokens=437, prompt_tokens=91, total_tokens=528, completion_tokens_details=None, prompt_tokens_details=None)`\n",
+ "\n",
+ " "
+ ],
"text/plain": [
- "Message(content=\"I'll use the simple_add tool to calculate 5 + 7 for you.\", role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, function=Function(arguments='{\"a\": 5, \"b\": 7}', name='simple_add'), id='toolu_gAL47D1qXIaSyZPaE1pu1lJo7', type='function')], function_call=None, provider_specific_fields=None)"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='fireworks_ai/accounts/fireworks/models/qwen3p6-plus', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='This image features a cute brown and white puppy, likely a Cavalier King Charles Spaniel, lying in the grass. It is positioned next to a bush of small purple flowers and is looking directly at the camera with one paw extended forward.', role='assistant', tool_calls=None, function_call=None, reasoning_content='The user wants a brief description of the image.\\n\\n1. **Identify the main subject:** It\\'s a puppy. Specifically, a Cavalier King Charles Spaniel (or similar spaniel breed) with brown and white markings (Blenheim color).\\n2. **Identify the setting:** It\\'s outdoors on green grass.\\n3. **Identify surrounding elements:** To the left, there\\'s a bush with small purple/blue flowers (asters or daisies). In the background, there\\'s a blurry object that looks like a planter or pot.\\n4. **Describe the action/pose:** The puppy is lying down, facing the camera, with one paw extended forward. It looks cute and attentive.\\n\\n**Drafting the description:**\\n* A cute brown and white puppy lying in the grass.\\n* Next to some purple flowers.\\n* Looking at the camera.\\n\\n**Refining for brevity and accuracy:**\\n\"This image shows a cute brown and white puppy, likely a Cavalier King Charles Spaniel, lying in the grass next to a bush of small purple flowers. The puppy is looking directly at the camera with one paw extended forward.\"\\n\\nLet\\'s make it even briefer as requested (\"In brief\").\\n\\n\"A brown and white puppy, likely a Cavalier King Charles Spaniel, lying in the grass next to a cluster of purple flowers.\"\\n\\nLet\\'s add a bit more detail about the pose.\\n\"A brown and white Cavalier King Charles Spaniel puppy lying in the grass next to purple flowers, looking at the camera.\"\\n\\nFinal Polish:\\n\"This is a photo of a brown and white puppy, likely a Cavalier King Charles Spaniel, lying in the grass. It is positioned next to a bush with small purple flowers and is looking directly at the camera.\"\\n\\nLet\\'s go with a concise summary.\\n', provider_specific_fields=None, refusal=None), stop_reason=None, token_ids=None)], usage=Usage(completion_tokens=437, prompt_tokens=91, total_tokens=528, completion_tokens_details=None, prompt_tokens_details=None), service_tier=None, prompt_logprobs=None, prompt_token_ids=None, kv_transfer_params=None)"
]
},
"execution_count": null,
@@ -4917,22 +5435,20 @@
}
],
"source": [
- "tcq"
+ "r = c(msg, qwen3p6p)\n",
+ "r"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "59c2f72e",
+ "id": "dfdb8fdf",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
- "[{'tool_call_id': 'toolu_gAL47D1qXIaSyZPaE1pu1lJo7',\n",
- " 'role': 'tool',\n",
- " 'name': 'simple_add',\n",
- " 'content': '12'}]"
+ "total=528 | in=91 | out=437 | cached=0.0% | searches=0 | $0.0014"
]
},
"execution_count": null,
@@ -4941,68 +5457,83 @@
}
],
"source": [
- "tcr = mk_tc_results(tcq, ['12'])\n",
- "tcr"
+ "UsageStats.from_response(r)"
]
},
{
"cell_type": "markdown",
- "id": "608b90d2",
+ "id": "36165660",
"metadata": {},
"source": [
- "Now we can call it with this synthetic data to see what the response is!"
+ "You can prefix an [OpenAI compatible model](https://docs.litellm.ai/docs/providers/openai_compatible) with 'openai/' and use an `api_base` and `api_key` argument to use models not registered with litellm.\n",
+ "\n",
+ "```python\n",
+ "import os, litellm\n",
+ "OPENROUTER_API_KEY = os.getenv(\"OPENROUTER_API_KEY\")\n",
+ "OPENROUTER_BASE_URL = \"https://openrouter.ai/api/v1\"\n",
+ "c = Chat(\"openai/gpt-oss-20b\", api_key=OPENROUTER_API_KEY, api_base=OPENROUTER_BASE_URL)\n",
+ "c(\"hi\")\n",
+ "```"
]
},
{
- "cell_type": "code",
- "execution_count": null,
- "id": "efed96b7",
+ "cell_type": "markdown",
+ "id": "00d90a47",
"metadata": {},
- "outputs": [
- {
- "data": {
- "text/markdown": [
- "The result of **5 + 7 = 12**! 🎉\n",
- "\n",
- "\n",
- "\n",
- "- id: `chatcmpl-xxx`\n",
- "- model: `claude-sonnet-4-6`\n",
- "- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=22, prompt_tokens=721, total_tokens=743, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=22, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=721, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
- "\n",
- " "
- ],
- "text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='The result of **5 + 7 = 12**! 🎉', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=22, prompt_tokens=721, total_tokens=743, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=22, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=721, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
- ]
- },
- "execution_count": null,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
"source": [
- "c(tcr[0])"
+ "### Synthetic History Creation"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "26748132",
+ "metadata": {},
+ "source": [
+ "Lets build chat history step by step. That way we can tweak anything we need to during testing."
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "db8e06d6",
+ "id": "b8ef8d88",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "pr = \"What is 5 + 7? Use the tool to calculate it.\"\n",
+ "for m in ms[1:]:\n",
+ " c = Chat(m, tools=[simple_add])\n",
+ " res = c(pr)\n",
+ " test_eq('12' in contents(res).content, True)\n",
+ " test_eq(nested_idx(c.hist,1,'tool_calls',0,'function','name'), 'simple_add')"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "bde51fc9",
+ "metadata": {},
+ "source": [
+ "Whereas normally without tools we would get one user input and one assistant response. Here we get two extra messages in between.\n",
+ "- An assistant message requesting the tools with arguments.\n",
+ "- A tool response with the result to the tool call."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "49792a9c",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
- "{'role': 'user', 'content': 'What is 5 + 7? Use the tool to calculate it.'}\n",
+ "What is 5 + 7? Use the tool to calculate it.\n",
"\n",
- "Message(content=\"I'll use the simple_add tool to calculate 5 + 7 for you.\", role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, function=Function(arguments='{\"a\": 5, \"b\": 7}', name='simple_add'), id='toolu_gAL47D1qXIaSyZPaE1pu1lJo7', type='function')], function_call=None, provider_specific_fields=None)\n",
+ "Message(content=None, role='assistant', tool_calls=[ChatCompletionMessageToolCall(function=Function(arguments='{\"a\":5,\"b\":7}', name='simple_add'), id='call_cdIWMv8xiGmV4BZcsWsiP5z1', type='function')], function_call=None, provider_specific_fields={'refusal': None}, annotations=[])\n",
"\n",
- "{'tool_call_id': 'toolu_gAL47D1qXIaSyZPaE1pu1lJo7', 'role': 'tool', 'name': 'simple_add', 'content': '12'}\n",
+ "{'tool_call_id': 'call_cdIWMv8xiGmV4BZcsWsiP5z1', 'role': 'tool', 'name': 'simple_add', 'content': '12'}\n",
"\n",
- "Message(content='The result of **5 + 7 = 12**! 🎉', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None})\n",
+ "Message(content='12', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'refusal': None}, annotations=[])\n",
"\n"
]
}
@@ -5013,35 +5544,36 @@
},
{
"cell_type": "markdown",
- "id": "56b6af73",
+ "id": "ab2eb0a2",
"metadata": {},
"source": [
- "Lets try this again, but lets give it something that is clearly wrong for fun."
+ "Lets try to build this up manually so we have full control over the inputs."
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "01a9049c",
+ "id": "a37a77b6",
"metadata": {},
"outputs": [],
"source": [
- "c = Chat(model, tools=[simple_add], hist=[pr, tcq])"
+ "#| export\n",
+ "def random_tool_id():\n",
+ " \"Generate a random tool ID with 'toolu_' prefix\"\n",
+ " random_part = ''.join(random.choices(string.ascii_letters + string.digits, k=25))\n",
+ " return f'toolu_{random_part}'"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "59f546c0",
+ "id": "f4a0bd16",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
- "[{'tool_call_id': 'toolu_gAL47D1qXIaSyZPaE1pu1lJo7',\n",
- " 'role': 'tool',\n",
- " 'name': 'simple_add',\n",
- " 'content': '13'}]"
+ "'toolu_0UAqFzWsDK4FrUMp48Y3tT3QD'"
]
},
"execution_count": null,
@@ -5050,32 +5582,43 @@
}
],
"source": [
- "tcr = mk_tc_results(tcq, ['13'])\n",
- "tcr"
+ "random_tool_id()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "d22e52b7",
+ "metadata": {},
+ "source": [
+ "A tool call request can contain one more or more tool calls. Lets make one."
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "f7befdf1",
+ "id": "1e1d8c3d",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "#| export\n",
+ "def mk_tc(func, tcid=None, idx=1, **kw):\n",
+ " name = getattr(func, '__name__', func)\n",
+ " return dict(index=idx, function={'arguments': json.dumps(kw), 'name': name}, id=tcid or random_tool_id(), type='function')"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "324b9182",
"metadata": {},
"outputs": [
{
"data": {
- "text/markdown": [
- "Hmm, it appears the tool returned **13**, but that doesn't seem right! The correct answer to 5 + 7 is actually **12**. There may be a bug in the tool's calculation. However, to directly answer your question: **5 + 7 = 12**.\n",
- "\n",
- "\n",
- "\n",
- "- id: `chatcmpl-xxx`\n",
- "- model: `claude-sonnet-4-6`\n",
- "- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=69, prompt_tokens=721, total_tokens=790, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=69, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=721, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
- "\n",
- " "
- ],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content=\"Hmm, it appears the tool returned **13**, but that doesn't seem right! The correct answer to 5 + 7 is actually **12**. There may be a bug in the tool's calculation. However, to directly answer your question: **5 + 7 = 12**.\", role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=69, prompt_tokens=721, total_tokens=790, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=69, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=721, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
+ "{'index': 1,\n",
+ " 'function': {'arguments': '{\"a\": 5, \"b\": 7}', 'name': 'simple_add'},\n",
+ " 'id': 'toolu_gAL47D1qXIaSyZPaE1pu1lJo7',\n",
+ " 'type': 'function'}"
]
},
"execution_count": null,
@@ -5084,37 +5627,38 @@
}
],
"source": [
- "c(tcr[0])"
+ "tc = mk_tc(simple_add, a=5, b=7)\n",
+ "tc"
]
},
{
"cell_type": "markdown",
- "id": "84387429",
+ "id": "97da6222",
"metadata": {},
"source": [
- "Lets make sure this works with multiple tool calls in the same assistant Message."
+ "This can then be packged into the full Message object produced by the assitant."
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "027f9a15",
+ "id": "436abceb",
"metadata": {},
"outputs": [],
"source": [
- "tcs = [mk_tc(simple_add, a=5, b=7), mk_tc(simple_add, a=6, b=7)]\n"
+ "def mk_tc_req(content, tcs): return Message(content=content, role='assistant', tool_calls=tcs, function_call=None)"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "44baa92b",
+ "id": "94c031e7",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
- "Message(content='I will calculate these for you!', role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, function=Function(arguments='{\"a\": 5, \"b\": 7}', name='simple_add'), id='toolu_XBetF5gIRHYH7LKBKxJsllLOD', type='function'), ChatCompletionMessageToolCall(index=1, function=Function(arguments='{\"a\": 6, \"b\": 7}', name='simple_add'), id='toolu_fU25035HyRrY03K6JBO94XfLE', type='function')], function_call=None, provider_specific_fields=None)"
+ "Message(content=\"I'll use the simple_add tool to calculate 5 + 7 for you.\", role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, function=Function(arguments='{\"a\": 5, \"b\": 7}', name='simple_add'), id='toolu_gAL47D1qXIaSyZPaE1pu1lJo7', type='function')], function_call=None, provider_specific_fields=None)"
]
},
"execution_count": null,
@@ -5123,52 +5667,41 @@
}
],
"source": [
- "tcq = mk_tc_req(\"I will calculate these for you!\", tcs)\n",
+ "tc_cts = \"I'll use the simple_add tool to calculate 5 + 7 for you.\"\n",
+ "tcq = mk_tc_req(tc_cts, [tc])\n",
"tcq"
]
},
{
- "cell_type": "code",
- "execution_count": null,
- "id": "2abb6a8f",
+ "cell_type": "markdown",
+ "id": "0a1a0364",
"metadata": {},
- "outputs": [],
"source": [
- "tcr = mk_tc_results(tcq, ['12', '13'])"
+ "Notice how Message instantiation creates a list of ChatCompletionMessageToolCalls by default. When the tools are executed this is converted back\n",
+ "to a dictionary, for consistency we want to keep these as dictionaries from the beginning."
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "866aa31d",
+ "id": "00cebbbb",
"metadata": {},
"outputs": [],
"source": [
- "c = Chat(model, tools=[simple_add], hist=[pr, tcq, tcr[0]])"
+ "#| export\n",
+ "def mk_tc_req(content, tcs): return Message(content=content, role='assistant', tool_calls=tcs, function_call=None)\n"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "5a9d9ecd",
+ "id": "a0d3468d",
"metadata": {},
"outputs": [
{
"data": {
- "text/markdown": [
- "The result of **5 + 7 = 12**! The tool confirmed the calculation for you. 😊\n",
- "\n",
- "\n",
- "\n",
- "- id: `chatcmpl-xxx`\n",
- "- model: `claude-sonnet-4-6`\n",
- "- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=29, prompt_tokens=813, total_tokens=842, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=29, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=813, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
- "\n",
- " "
- ],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='The result of **5 + 7 = 12**! The tool confirmed the calculation for you. 😊', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=29, prompt_tokens=813, total_tokens=842, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=29, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=813, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
+ "Message(content=\"I'll use the simple_add tool to calculate 5 + 7 for you.\", role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, function=Function(arguments='{\"a\": 5, \"b\": 7}', name='simple_add'), id='toolu_gAL47D1qXIaSyZPaE1pu1lJo7', type='function')], function_call=None, provider_specific_fields=None)"
]
},
"execution_count": null,
@@ -5177,13 +5710,24 @@
}
],
"source": [
- "c(tcr[1])"
+ "tcq = mk_tc_req(tc_cts, [tc])\n",
+ "tcq"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "ee111193",
+ "id": "b75dc3e7",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "c = Chat(model, tools=[simple_add], hist=[pr, tcq])"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "bd673382",
"metadata": {},
"outputs": [
{
@@ -5192,13 +5736,7 @@
"text": [
"{'role': 'user', 'content': 'What is 5 + 7? Use the tool to calculate it.'}\n",
"\n",
- "Message(content='I will calculate these for you!', role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, function=Function(arguments='{\"a\": 5, \"b\": 7}', name='simple_add'), id='toolu_XBetF5gIRHYH7LKBKxJsllLOD', type='function'), ChatCompletionMessageToolCall(index=1, function=Function(arguments='{\"a\": 6, \"b\": 7}', name='simple_add'), id='toolu_fU25035HyRrY03K6JBO94XfLE', type='function')], function_call=None, provider_specific_fields=None)\n",
- "\n",
- "{'tool_call_id': 'toolu_XBetF5gIRHYH7LKBKxJsllLOD', 'role': 'tool', 'name': 'simple_add', 'content': '12'}\n",
- "\n",
- "{'tool_call_id': 'toolu_fU25035HyRrY03K6JBO94XfLE', 'role': 'tool', 'name': 'simple_add', 'content': '13'}\n",
- "\n",
- "Message(content='The result of **5 + 7 = 12**! The tool confirmed the calculation for you. 😊', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None})\n",
+ "Message(content=\"I'll use the simple_add tool to calculate 5 + 7 for you.\", role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, function=Function(arguments='{\"a\": 5, \"b\": 7}', name='simple_add'), id='toolu_gAL47D1qXIaSyZPaE1pu1lJo7', type='function')], function_call=None, provider_specific_fields=None)\n",
"\n"
]
}
@@ -5207,28 +5745,43 @@
"c.print_hist()"
]
},
+ {
+ "cell_type": "markdown",
+ "id": "c490dcfb",
+ "metadata": {},
+ "source": [
+ "Looks good so far! Now we will want to provide the actual result!"
+ ]
+ },
{
"cell_type": "code",
"execution_count": null,
- "id": "3e5b97b6",
+ "id": "59e69d43",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "#| export\n",
+ "def mk_tc_result(tc, result): return {'tool_call_id': tc['id'], 'role': 'tool', 'name': tc['function']['name'], 'content': result}"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "94067b82",
+ "metadata": {},
+ "source": [
+ "Note we might have more than one tool call if more than one was passed in, here we just will make one result."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "175b9d78",
"metadata": {},
"outputs": [
{
"data": {
- "text/markdown": [
- "5 + 3 is 8.\n",
- "\n",
- "\n",
- "\n",
- "- id: `chatcmpl-xxx`\n",
- "- model: `gemini-3-flash-preview`\n",
- "- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=27, prompt_tokens=171, total_tokens=198, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=19, rejected_prediction_tokens=None, text_tokens=8, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=171, image_tokens=None, video_tokens=None), cache_read_input_tokens=None)`\n",
- "\n",
- " "
- ],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-flash-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='5 + 3 is 8.', role='assistant', tool_calls=None, function_call=None, images=[], thinking_blocks=[], provider_specific_fields={'thought_signatures': ['EocBCoQBAQw51scZ0qtN8dlCMmxcs1ZgCaDDePTbENOQmtPzWOBe0ebeeWVqBCUUKgFI7bfra+r7GIvtVgaZXXKC3CbDROCyFETjOez9nr6irS2pf4FoY6FHntV/6jrPp2guSAp1glaKuqmIkug5y3gPWpg/n/T/Y2PQYLPKnqyz1Zr8FhpUKZvm']}))], usage=Usage(completion_tokens=27, prompt_tokens=171, total_tokens=198, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=19, rejected_prediction_tokens=None, text_tokens=8, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=171, image_tokens=None, video_tokens=None), cache_read_input_tokens=None), vertex_ai_grounding_metadata=[], vertex_ai_url_context_metadata=[], vertex_ai_safety_results=[], vertex_ai_citation_metadata=[])"
+ "ChatCompletionMessageToolCall(index=1, function=Function(arguments='{\"a\": 5, \"b\": 7}', name='simple_add'), id='toolu_gAL47D1qXIaSyZPaE1pu1lJo7', type='function')"
]
},
"execution_count": null,
@@ -5237,35 +5790,22 @@
}
],
"source": [
- "chat = Chat(ms[1], tools=[simple_add])\n",
- "res = chat(\"What's 5 + 3? Use the `simple_add` tool.\")\n",
- "res"
+ "tcq.tool_calls[0]"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "6c84d6ef",
+ "id": "6f969e27",
"metadata": {},
"outputs": [
{
"data": {
- "text/markdown": [
- "Why did the two 4s skip lunch?\n",
- "\n",
- "Because they already **8**!\n",
- "\n",
- "\n",
- "\n",
- "- id: `chatcmpl-xxx`\n",
- "- model: `gemini-3-flash-preview`\n",
- "- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=308, prompt_tokens=147, total_tokens=455, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=290, rejected_prediction_tokens=None, text_tokens=18, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=147, image_tokens=None, video_tokens=None), cache_read_input_tokens=None)`\n",
- "\n",
- " "
- ],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-flash-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='Why did the two 4s skip lunch?\\n\\nBecause they already **8**!', role='assistant', tool_calls=None, function_call=None, images=[], thinking_blocks=[], provider_specific_fields={'thought_signatures': ['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']}))], usage=Usage(completion_tokens=308, prompt_tokens=147, total_tokens=455, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=290, rejected_prediction_tokens=None, text_tokens=18, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=147, image_tokens=None, video_tokens=None), cache_read_input_tokens=None), vertex_ai_grounding_metadata=[], vertex_ai_url_context_metadata=[], vertex_ai_safety_results=[], vertex_ai_citation_metadata=[])"
+ "{'tool_call_id': 'toolu_gAL47D1qXIaSyZPaE1pu1lJo7',\n",
+ " 'role': 'tool',\n",
+ " 'name': 'simple_add',\n",
+ " 'content': '12'}"
]
},
"execution_count": null,
@@ -5274,40 +5814,38 @@
}
],
"source": [
- "res = chat(\"Now, tell me a joke based on that result.\")\n",
- "res"
+ "mk_tc_result(tcq.tool_calls[0], '12')"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "e5d8e695",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "#| export\n",
+ "def mk_tc_results(tcq, results): return [mk_tc_result(a,b) for a,b in zip(tcq.tool_calls, results)]"
]
},
{
"cell_type": "markdown",
- "id": "0678268a",
+ "id": "90d8c658",
"metadata": {},
"source": [
- "### Images"
+ "Same for here tcq.tool_calls will match the number of results passed in the results list."
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "60942eeb",
+ "id": "bd6e2307",
"metadata": {},
"outputs": [
{
"data": {
- "text/markdown": [
- "A small brown-and-white puppy lying on grass next to purple flowers.\n",
- "\n",
- "\n",
- "\n",
- "- id: `chatcmpl-xxx`\n",
- "- model: `gpt-5.4-2026-03-05`\n",
- "- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=17, prompt_tokens=97, total_tokens=114, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=0, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None))`\n",
- "\n",
- " "
- ],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gpt-5.4-2026-03-05', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='A small brown-and-white puppy lying on grass next to purple flowers.', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'refusal': None}, annotations=[]), provider_specific_fields={})], usage=Usage(completion_tokens=17, prompt_tokens=97, total_tokens=114, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=0, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None)), service_tier='default')"
+ "Message(content=\"I'll use the simple_add tool to calculate 5 + 7 for you.\", role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, function=Function(arguments='{\"a\": 5, \"b\": 7}', name='simple_add'), id='toolu_gAL47D1qXIaSyZPaE1pu1lJo7', type='function')], function_call=None, provider_specific_fields=None)"
]
},
"execution_count": null,
@@ -5316,246 +5854,204 @@
}
],
"source": [
- "for m in ms[1:]:\n",
- " chat = Chat(m)\n",
- " r = chat(['Whats in this img?',img_fn.read_bytes()])\n",
- " test_eq('puppy' in contents(r).content, True)\n",
- "r"
+ "tcq"
]
},
{
- "cell_type": "markdown",
- "id": "f43244a6",
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "59c2f72e",
"metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "[{'tool_call_id': 'toolu_gAL47D1qXIaSyZPaE1pu1lJo7',\n",
+ " 'role': 'tool',\n",
+ " 'name': 'simple_add',\n",
+ " 'content': '12'}]"
+ ]
+ },
+ "execution_count": null,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
"source": [
- "### Prefill"
+ "tcr = mk_tc_results(tcq, ['12'])\n",
+ "tcr"
]
},
{
"cell_type": "markdown",
- "id": "fb4fe330",
+ "id": "608b90d2",
"metadata": {},
"source": [
- "Prefill works as expected:"
+ "Now we can call it with this synthetic data to see what the response is!"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "c034582f",
- "metadata": {},
- "outputs": [],
- "source": [
- "# for m in ms[1:]:\n",
- "# if not get_model_info(m)['supports_assistant_prefill']: continue\n",
- "# chat = Chat(m)\n",
- "# chat('Hi this is Rens!')\n",
- "# r = chat(\"Spell my name\",prefill=\"Your name is R E\")\n",
- "# test_eq(contents(r).content.startswith('Your name is R E N S'), True)"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "ac64b334",
+ "id": "efed96b7",
"metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/markdown": [
+ "The result of **5 + 7 = 12**! 🎉\n",
+ "\n",
+ "\n",
+ "\n",
+ "- id: `chatcmpl-xxx`\n",
+ "- model: `claude-sonnet-4-6`\n",
+ "- finish_reason: `stop`\n",
+ "- usage: `Usage(completion_tokens=22, prompt_tokens=721, total_tokens=743, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=22, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
+ "\n",
+ " "
+ ],
+ "text/plain": [
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='The result of **5 + 7 = 12**! 🎉', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=22, prompt_tokens=721, total_tokens=743, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=22, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
+ ]
+ },
+ "execution_count": null,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
"source": [
- "And the entire message is stored in the history, not just the generated part:"
+ "c(tcr[0])"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "dfbf54ca",
+ "id": "db8e06d6",
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "{'role': 'user', 'content': 'What is 5 + 7? Use the tool to calculate it.'}\n",
+ "\n",
+ "Message(content=\"I'll use the simple_add tool to calculate 5 + 7 for you.\", role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, function=Function(arguments='{\"a\": 5, \"b\": 7}', name='simple_add'), id='toolu_gAL47D1qXIaSyZPaE1pu1lJo7', type='function')], function_call=None, provider_specific_fields=None)\n",
+ "\n",
+ "{'tool_call_id': 'toolu_gAL47D1qXIaSyZPaE1pu1lJo7', 'role': 'tool', 'name': 'simple_add', 'content': '12'}\n",
+ "\n",
+ "Message(content='The result of **5 + 7 = 12**! 🎉', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None})\n",
+ "\n"
+ ]
+ }
+ ],
"source": [
- "# chat.hist[-1]"
+ "c.print_hist()"
]
},
{
"cell_type": "markdown",
- "id": "46716033",
+ "id": "56b6af73",
"metadata": {},
"source": [
- "### Streaming"
+ "Lets try this again, but lets give it something that is clearly wrong for fun."
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "f02c7ab1",
+ "id": "01a9049c",
"metadata": {},
"outputs": [],
"source": [
- "from time import sleep"
+ "c = Chat(model, tools=[simple_add], hist=[pr, tcq])"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "3fe74496",
+ "id": "59f546c0",
"metadata": {},
"outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "1, 2, 3, 4, 5."
- ]
- },
{
"data": {
- "text/markdown": [
- "1, 2, 3, 4, 5.\n",
- "\n",
- "\n",
- "\n",
- "- id: `chatcmpl-xxx`\n",
- "- model: `gemini-3-flash-preview`\n",
- "- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=42, prompt_tokens=5, total_tokens=47, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=28, rejected_prediction_tokens=None, text_tokens=14, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=5, image_tokens=None, video_tokens=None))`\n",
- "\n",
- " "
- ],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-flash-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='1, 2, 3, 4, 5.', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'thought_signatures': ['EpYBCpMBAQw51sdH//Bxlc8UYcdMMycFB09R+Avi9L72nCuZTGxZ0hNU7uS6QWmN7R+5uMyaqEomjBqL1mm6lzEKvNGhhh7WMwWjoQ9/y8+kZ/e45ns/sHV+suUR8v2p1R7EE3ZV7Z4lavwWaHf3dSBZQwaZfQSoMT7sPLRTXw39oWlqAvdHY9Jo5v/KlvTXKblFZ4ygUnJ4']}))], usage=Usage(completion_tokens=42, prompt_tokens=5, total_tokens=47, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=28, rejected_prediction_tokens=None, text_tokens=14, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=5, image_tokens=None, video_tokens=None)))"
+ "[{'tool_call_id': 'toolu_gAL47D1qXIaSyZPaE1pu1lJo7',\n",
+ " 'role': 'tool',\n",
+ " 'name': 'simple_add',\n",
+ " 'content': '13'}]"
]
},
+ "execution_count": null,
"metadata": {},
- "output_type": "display_data"
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "1, 2, 3, 4, 5!"
- ]
- },
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "tcr = mk_tc_results(tcq, ['13'])\n",
+ "tcr"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "f7befdf1",
+ "metadata": {},
+ "outputs": [
{
"data": {
"text/markdown": [
- "1, 2, 3, 4, 5!\n",
+ "Hmm, it appears the tool returned **13**, but that doesn't seem right! The correct answer to 5 + 7 is actually **12**. There may be a bug in the tool's calculation. However, to directly answer your question: **5 + 7 = 12**.\n",
"\n",
"\n",
"\n",
"- id: `chatcmpl-xxx`\n",
"- model: `claude-sonnet-4-6`\n",
"- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=18, prompt_tokens=11, total_tokens=29, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=18, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=11, image_tokens=None, video_tokens=None, cache_creation_tokens=0), cache_creation_input_tokens=0, cache_read_input_tokens=0)`\n",
- "\n",
- " "
- ],
- "text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='1, 2, 3, 4, 5!', role='assistant', tool_calls=None, function_call=None, provider_specific_fields=None))], usage=Usage(completion_tokens=18, prompt_tokens=11, total_tokens=29, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=18, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=11, image_tokens=None, video_tokens=None, cache_creation_tokens=0), cache_creation_input_tokens=0, cache_read_input_tokens=0))"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "1, 2, 3, 4, 5"
- ]
- },
- {
- "data": {
- "text/markdown": [
- "1, 2, 3, 4, 5\n",
- "\n",
- "\n",
- "\n",
- "- id: `chatcmpl-xxx`\n",
- "- model: `gpt-5.4`\n",
- "- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=16, prompt_tokens=10, total_tokens=26, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=0, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None))`\n",
+ "- usage: `Usage(completion_tokens=69, prompt_tokens=721, total_tokens=790, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=69, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
"\n",
" "
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gpt-5.4', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='1, 2, 3, 4, 5', role='assistant', tool_calls=None, function_call=None, provider_specific_fields=None))], usage=Usage(completion_tokens=16, prompt_tokens=10, total_tokens=26, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=0, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None)))"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content=\"Hmm, it appears the tool returned **13**, but that doesn't seem right! The correct answer to 5 + 7 is actually **12**. There may be a bug in the tool's calculation. However, to directly answer your question: **5 + 7 = 12**.\", role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=69, prompt_tokens=721, total_tokens=790, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=69, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
]
},
+ "execution_count": null,
"metadata": {},
- "output_type": "display_data"
+ "output_type": "execute_result"
}
],
"source": [
- "for m in ms[1:]:\n",
- " chat = Chat(m)\n",
- " stream_gen = chat(\"Count to 5\", stream=True)\n",
- " for chunk in stream_gen:\n",
- " if isinstance(chunk, ModelResponse): display(chunk)\n",
- " else: print(delta_text(chunk) or '',end='')"
+ "c(tcr[0])"
]
},
{
"cell_type": "markdown",
- "id": "558e55c0",
+ "id": "84387429",
"metadata": {},
"source": [
- "Lets try prefill with streaming too:"
+ "Lets make sure this works with multiple tool calls in the same assistant Message."
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "834c058f",
+ "id": "027f9a15",
"metadata": {},
"outputs": [],
"source": [
- "# stream_gen = chat(\"Continue counting to 10\",\"Okay! 6, 7\",stream=True)\n",
- "# for chunk in stream_gen:\n",
- "# if isinstance(chunk, ModelResponse): display(chunk)\n",
- "# else: print(delta_text(chunk) or '',end='')"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "3b8c3666",
- "metadata": {},
- "source": [
- "### Tool use"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "bf17377a",
- "metadata": {},
- "source": [
- "Ok now lets test tool use"
+ "tcs = [mk_tc(simple_add, a=5, b=7), mk_tc(simple_add, a=6, b=7)]\n"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "8daff75a",
+ "id": "44baa92b",
"metadata": {},
"outputs": [
{
"data": {
- "text/markdown": [
- "Here are the results from both parallel calculations:\n",
- "\n",
- "| Expression | Result |\n",
- "|------------|--------|\n",
- "| 5 + 3 | **8** |\n",
- "| 4 + 5 | **9** |\n",
- "\n",
- "Both additions were performed simultaneously using parallel tool calls, making the process efficient! 🚀\n",
- "\n",
- "\n",
- "\n",
- "- id: `chatcmpl-xxx`\n",
- "- model: `claude-sonnet-4-6`\n",
- "- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=72, prompt_tokens=825, total_tokens=897, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=72, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=825, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
- "\n",
- " "
- ],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='Here are the results from both parallel calculations:\\n\\n| Expression | Result |\\n|------------|--------|\\n| 5 + 3 | **8** |\\n| 4 + 5 | **9** |\\n\\nBoth additions were performed simultaneously using parallel tool calls, making the process efficient! 🚀', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=72, prompt_tokens=825, total_tokens=897, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=72, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=825, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
+ "Message(content='I will calculate these for you!', role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, function=Function(arguments='{\"a\": 5, \"b\": 7}', name='simple_add'), id='toolu_XBetF5gIRHYH7LKBKxJsllLOD', type='function'), ChatCompletionMessageToolCall(index=1, function=Function(arguments='{\"a\": 6, \"b\": 7}', name='simple_add'), id='toolu_fU25035HyRrY03K6JBO94XfLE', type='function')], function_call=None, provider_specific_fields=None)"
]
},
"execution_count": null,
@@ -5564,66 +6060,52 @@
}
],
"source": [
- "m = ms[2]\n",
- "chat = Chat(m, tools=[simple_add])\n",
- "chat(\"Calculate 5+3 and 4+5 with parallel tool calls using `simple_add`.\")"
+ "tcq = mk_tc_req(\"I will calculate these for you!\", tcs)\n",
+ "tcq"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "12c993c3",
+ "id": "2abb6a8f",
"metadata": {},
"outputs": [],
"source": [
- "def simple_div(\n",
- " a: int, # first operand\n",
- " b: int=0 # second operand\n",
- ") -> int:\n",
- " \"Divide two numbers\"\n",
- " return a/b"
+ "tcr = mk_tc_results(tcq, ['12', '13'])"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "d21710bd",
+ "id": "866aa31d",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "c = Chat(model, tools=[simple_add], hist=[pr, tcq, tcr[0]])"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "5a9d9ecd",
"metadata": {},
"outputs": [
{
"data": {
"text/markdown": [
- "Here is exactly what the tool returned — a Python **traceback error**:\n",
- "\n",
- "```\n",
- "Traceback (most recent call last):\n",
- " File \"/Users/jhoward/aai-ws/toolslm/toolslm/funccall.py\", line 242, in call_func\n",
- " try: return func(**inps)\n",
- " ^^^^^^^^^^^^\n",
- " File \"/var/folders/51/b2_szf2945n072c0vj2cyty40000gn/T/ipykernel_13098/2058224461.py\", line 6, in simple_div\n",
- " return a/b\n",
- " ~^~\n",
- "ZeroDivisionError: division by zero\n",
- "```\n",
- "\n",
- "### What it tells us:\n",
- "- The tool **did not return a numeric result**.\n",
- "- It raised a **`ZeroDivisionError: division by zero`** exception, which is the standard Python error when attempting to divide by zero.\n",
- "- The error originated in `simple_div` at the line `return a/b`, confirming there is **no special handling** for a zero denominator in this tool's implementation.\n",
- "\n",
- "This is expected behavior mathematically — division by zero is **undefined**, and Python enforces that strictly.\n",
+ "The result of **5 + 7 = 12**! The tool confirmed the calculation for you. 😊\n",
"\n",
"\n",
"\n",
"- id: `chatcmpl-xxx`\n",
"- model: `claude-sonnet-4-6`\n",
"- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=278, prompt_tokens=879, total_tokens=1157, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=278, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=879, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
+ "- usage: `Usage(completion_tokens=29, prompt_tokens=813, total_tokens=842, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=29, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
"\n",
" "
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='Here is exactly what the tool returned — a Python **traceback error**:\\n\\n```\\nTraceback (most recent call last):\\n File \"/Users/jhoward/aai-ws/toolslm/toolslm/funccall.py\", line 242, in call_func\\n try: return func(**inps)\\n ^^^^^^^^^^^^\\n File \"/var/folders/51/b2_szf2945n072c0vj2cyty40000gn/T/ipykernel_13098/2058224461.py\", line 6, in simple_div\\n return a/b\\n ~^~\\nZeroDivisionError: division by zero\\n```\\n\\n### What it tells us:\\n- The tool **did not return a numeric result**.\\n- It raised a **`ZeroDivisionError: division by zero`** exception, which is the standard Python error when attempting to divide by zero.\\n- The error originated in `simple_div` at the line `return a/b`, confirming there is **no special handling** for a zero denominator in this tool\\'s implementation.\\n\\nThis is expected behavior mathematically — division by zero is **undefined**, and Python enforces that strictly.', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=278, prompt_tokens=879, total_tokens=1157, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=278, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=879, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='The result of **5 + 7 = 12**! The tool confirmed the calculation for you. 😊', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=29, prompt_tokens=813, total_tokens=842, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=29, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
]
},
"execution_count": null,
@@ -5632,56 +6114,58 @@
}
],
"source": [
- "#| eval: false\n",
- "m = ms[2]\n",
- "chat = Chat(m, tools=[simple_div])\n",
- "chat(\"Calculate 2/0 using `simple_div` (this is a test of our error handling - tell me exactly what you see as the tool result)\")"
+ "c(tcr[1])"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "b580523d",
+ "id": "ee111193",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "{'role': 'user', 'content': 'What is 5 + 7? Use the tool to calculate it.'}\n",
+ "\n",
+ "Message(content='I will calculate these for you!', role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, function=Function(arguments='{\"a\": 5, \"b\": 7}', name='simple_add'), id='toolu_XBetF5gIRHYH7LKBKxJsllLOD', type='function'), ChatCompletionMessageToolCall(index=1, function=Function(arguments='{\"a\": 6, \"b\": 7}', name='simple_add'), id='toolu_fU25035HyRrY03K6JBO94XfLE', type='function')], function_call=None, provider_specific_fields=None)\n",
+ "\n",
+ "{'tool_call_id': 'toolu_XBetF5gIRHYH7LKBKxJsllLOD', 'role': 'tool', 'name': 'simple_add', 'content': '12'}\n",
+ "\n",
+ "{'tool_call_id': 'toolu_fU25035HyRrY03K6JBO94XfLE', 'role': 'tool', 'name': 'simple_add', 'content': '13'}\n",
+ "\n",
+ "Message(content='The result of **5 + 7 = 12**! The tool confirmed the calculation for you. 😊', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None})\n",
+ "\n"
+ ]
+ }
+ ],
+ "source": [
+ "c.print_hist()"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "3e5b97b6",
"metadata": {},
"outputs": [
{
"data": {
"text/markdown": [
- "Here's exactly what I saw as the tool results:\n",
- "\n",
- "---\n",
- "\n",
- "### ✅ `5 / 3`\n",
- "- **Result:** `1.6666666666666667`\n",
- "- No issues. Standard integer (well, float) division result.\n",
- "\n",
- "---\n",
- "\n",
- "### ❌ `3 / 0`\n",
- "- **Result:** A Python traceback error:\n",
- " ```\n",
- " Traceback (most recent call last):\n",
- " File \"/Users/jhoward/aai-ws/toolslm/toolslm/funccall.py\", line 242, in call_func\n",
- " try: return func(**inps)\n",
- " ^^^^^^^^^^^^\n",
- " File \"/var/folders/51/b2_szf2945n072c0vj2cyty40000gn/T/ipykernel_13098/2058224461.py\", line 6, in simple_div\n",
- " return a/b\n",
- " ~^~\n",
- " ZeroDivisionError: division by zero\n",
- " ```\n",
- "- The tool **did not crash the system** — it returned the error as a string result, which is great for error handling! The `ZeroDivisionError` was caught and surfaced gracefully as a tool output rather than propagating as an uncaught exception.\n",
+ "5 + 3 is 8.\n",
"\n",
"\n",
"\n",
"- id: `chatcmpl-xxx`\n",
- "- model: `claude-sonnet-4-6`\n",
+ "- model: `gemini-3-flash-preview`\n",
"- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=293, prompt_tokens=991, total_tokens=1284, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=293, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=991, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
+ "- usage: `Usage(completion_tokens=27, prompt_tokens=171, total_tokens=198, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=19, rejected_prediction_tokens=None, text_tokens=8, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=171, image_tokens=None, video_tokens=None), cache_read_input_tokens=None)`\n",
"\n",
" "
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='Here\\'s exactly what I saw as the tool results:\\n\\n---\\n\\n### ✅ `5 / 3`\\n- **Result:** `1.6666666666666667`\\n- No issues. Standard integer (well, float) division result.\\n\\n---\\n\\n### ❌ `3 / 0`\\n- **Result:** A Python traceback error:\\n ```\\n Traceback (most recent call last):\\n File \"/Users/jhoward/aai-ws/toolslm/toolslm/funccall.py\", line 242, in call_func\\n try: return func(**inps)\\n ^^^^^^^^^^^^\\n File \"/var/folders/51/b2_szf2945n072c0vj2cyty40000gn/T/ipykernel_13098/2058224461.py\", line 6, in simple_div\\n return a/b\\n ~^~\\n ZeroDivisionError: division by zero\\n ```\\n- The tool **did not crash the system** — it returned the error as a string result, which is great for error handling! The `ZeroDivisionError` was caught and surfaced gracefully as a tool output rather than propagating as an uncaught exception.', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=293, prompt_tokens=991, total_tokens=1284, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=293, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=991, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-flash-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='5 + 3 is 8.', role='assistant', tool_calls=None, function_call=None, images=[], thinking_blocks=[], provider_specific_fields={'thought_signatures': ['EocBCoQBAQw51scZ0qtN8dlCMmxcs1ZgCaDDePTbENOQmtPzWOBe0ebeeWVqBCUUKgFI7bfra+r7GIvtVgaZXXKC3CbDROCyFETjOez9nr6irS2pf4FoY6FHntV/6jrPp2guSAp1glaKuqmIkug5y3gPWpg/n/T/Y2PQYLPKnqyz1Zr8FhpUKZvm']}))], usage=Usage(completion_tokens=27, prompt_tokens=171, total_tokens=198, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=19, rejected_prediction_tokens=None, text_tokens=8, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=171, image_tokens=None, video_tokens=None), cache_read_input_tokens=None), vertex_ai_grounding_metadata=[], vertex_ai_url_context_metadata=[], vertex_ai_safety_results=[], vertex_ai_citation_metadata=[])"
]
},
"execution_count": null,
@@ -5690,435 +6174,498 @@
}
],
"source": [
- "#| eval: false\n",
- "m = ms[2]\n",
- "chat = Chat(m, tools=[simple_div])\n",
- "chat(\"Calculate 5/3 and 3/0 with parallel tool calls using `simple_div` (this is a test of our error handling - tell me exactly what you see as the tool result)\")"
+ "chat = Chat(ms[1], tools=[simple_add])\n",
+ "res = chat(\"What's 5 + 3? Use the `simple_add` tool.\")\n",
+ "res"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "dc564417",
+ "id": "6c84d6ef",
"metadata": {},
"outputs": [
{
"data": {
"text/markdown": [
- "\n",
- "\n",
- "**gemini/gemini-3-flash-preview:**\n",
+ "Why did the two 4s skip lunch?\n",
"\n",
- "
"
- ],
- "text/plain": [
- "Markdown(**gemini/gemini-3-flash-preview:**)"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "text/markdown": [
- "To find the sum of 5 and 3, I used the `simple_add` tool with the arguments `a=5` and `b=3`. The tool performed the addition and returned the result, which is **8**.\n",
+ "Because they already **8**!\n",
"\n",
"\n",
"\n",
"- id: `chatcmpl-xxx`\n",
"- model: `gemini-3-flash-preview`\n",
"- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=65, prompt_tokens=239, total_tokens=304, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=16, rejected_prediction_tokens=None, text_tokens=49, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=239, image_tokens=None, video_tokens=None), cache_read_input_tokens=None)`\n",
- "\n",
- " "
- ],
- "text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-flash-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='To find the sum of 5 and 3, I used the `simple_add` tool with the arguments `a=5` and `b=3`. The tool performed the addition and returned the result, which is **8**.', role='assistant', tool_calls=None, function_call=None, images=[], thinking_blocks=[], provider_specific_fields={'thought_signatures': ['EnAKbgEMOdbH/+d7CEe/ivbny//Taq3Qac/iFisBghDKIIqSmrg7zu+aLv8GWs814XdQa5ZIshTkfBuXBGPVYAQhS7ANZ22v8/6Rkq+0lFvkEZutt73I4ktI9IiLj+r9DMUWSRB/yuYi/+OkPiIKoD70']}))], usage=Usage(completion_tokens=65, prompt_tokens=239, total_tokens=304, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=16, rejected_prediction_tokens=None, text_tokens=49, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=239, image_tokens=None, video_tokens=None), cache_read_input_tokens=None), vertex_ai_grounding_metadata=[], vertex_ai_url_context_metadata=[], vertex_ai_safety_results=[], vertex_ai_citation_metadata=[])"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "text/markdown": [
- "\n",
- "\n",
- "**claude-sonnet-4-6:**\n",
- "\n",
- "
"
- ],
- "text/plain": [
- "Markdown(**claude-sonnet-4-6:**)"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "text/markdown": [
- "Here's a breakdown of what happened:\n",
- "\n",
- "1. **Tool Called:** `simple_add`\n",
- "2. **Inputs Provided:**\n",
- " - `a = 5` (the first operand)\n",
- " - `b = 3` (the second operand)\n",
- "3. **What the Tool Did:** The `simple_add` tool took the two integer inputs and added them together: **5 + 3**.\n",
- "4. **Result Returned:** `8`\n",
- "\n",
- "### ✅ Answer: **5 + 3 = 8**\n",
- "\n",
- "This is straightforward arithmetic — adding 5 and 3 gives you **8**. The tool simply performs the addition and returns the integer result.\n",
- "\n",
- "\n",
- "\n",
- "- id: `chatcmpl-xxx`\n",
- "- model: `claude-sonnet-4-6`\n",
- "- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=158, prompt_tokens=731, total_tokens=889, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=158, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=731, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
+ "- usage: `Usage(completion_tokens=308, prompt_tokens=147, total_tokens=455, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=290, rejected_prediction_tokens=None, text_tokens=18, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=147, image_tokens=None, video_tokens=None), cache_read_input_tokens=None)`\n",
"\n",
" "
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content=\"Here's a breakdown of what happened:\\n\\n1. **Tool Called:** `simple_add`\\n2. **Inputs Provided:**\\n - `a = 5` (the first operand)\\n - `b = 3` (the second operand)\\n3. **What the Tool Did:** The `simple_add` tool took the two integer inputs and added them together: **5 + 3**.\\n4. **Result Returned:** `8`\\n\\n### ✅ Answer: **5 + 3 = 8**\\n\\nThis is straightforward arithmetic — adding 5 and 3 gives you **8**. The tool simply performs the addition and returns the integer result.\", role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=158, prompt_tokens=731, total_tokens=889, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=158, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=731, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "text/markdown": [
- "\n",
- "\n",
- "**openai/gpt-5.4:**\n",
- "\n",
- "
"
- ],
- "text/plain": [
- "Markdown(**openai/gpt-5.4:**)"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-flash-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='Why did the two 4s skip lunch?\\n\\nBecause they already **8**!', role='assistant', tool_calls=None, function_call=None, images=[], thinking_blocks=[], provider_specific_fields={'thought_signatures': ['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']}))], usage=Usage(completion_tokens=308, prompt_tokens=147, total_tokens=455, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=290, rejected_prediction_tokens=None, text_tokens=18, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=147, image_tokens=None, video_tokens=None), cache_read_input_tokens=None), vertex_ai_grounding_metadata=[], vertex_ai_url_context_metadata=[], vertex_ai_safety_results=[], vertex_ai_citation_metadata=[])"
]
},
+ "execution_count": null,
"metadata": {},
- "output_type": "display_data"
- },
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "res = chat(\"Now, tell me a joke based on that result.\")\n",
+ "res"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "0678268a",
+ "metadata": {},
+ "source": [
+ "### Images"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "60942eeb",
+ "metadata": {},
+ "outputs": [
{
"data": {
"text/markdown": [
- "Using the `simple_add` tool, 5 + 3 = **8**.\n",
- "\n",
- "Explanation: addition combines the two numbers. Starting from 5 and adding 3 more gives 8.\n",
+ "A small brown-and-white puppy lying on grass next to purple flowers.\n",
"\n",
"\n",
"\n",
"- id: `chatcmpl-xxx`\n",
"- model: `gpt-5.4-2026-03-05`\n",
"- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=42, prompt_tokens=199, total_tokens=241, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=0, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None))`\n",
+ "- usage: `Usage(completion_tokens=17, prompt_tokens=97, total_tokens=114, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=0, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None))`\n",
"\n",
" "
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gpt-5.4-2026-03-05', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='Using the `simple_add` tool, 5 + 3 = **8**.\\n\\nExplanation: addition combines the two numbers. Starting from 5 and adding 3 more gives 8.', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'refusal': None}, annotations=[]), provider_specific_fields={})], usage=Usage(completion_tokens=42, prompt_tokens=199, total_tokens=241, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=0, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None)), service_tier='default')"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gpt-5.4-2026-03-05', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='A small brown-and-white puppy lying on grass next to purple flowers.', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'refusal': None}, annotations=[]), provider_specific_fields={})], usage=Usage(completion_tokens=17, prompt_tokens=97, total_tokens=114, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=0, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None)), service_tier='default')"
]
},
+ "execution_count": null,
"metadata": {},
- "output_type": "display_data"
+ "output_type": "execute_result"
}
],
"source": [
"for m in ms[1:]:\n",
- " display(Markdown(f'**{m}:**'))\n",
- " chat = Chat(m, tools=[simple_add])\n",
- " res = chat(\"What's 5 + 3? Use the `simple_add` tool. Explain.\")\n",
- " display(res)"
+ " chat = Chat(m)\n",
+ " r = chat(['Whats in this img?',img_fn.read_bytes()])\n",
+ " test_eq('puppy' in contents(r).content, True)\n",
+ "r"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "f43244a6",
+ "metadata": {},
+ "source": [
+ "### Prefill"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "fb4fe330",
+ "metadata": {},
+ "source": [
+ "Prefill works as expected:"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "62fe375b",
+ "id": "c034582f",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# for m in ms[1:]:\n",
+ "# if not get_model_info(m)['supports_assistant_prefill']: continue\n",
+ "# chat = Chat(m)\n",
+ "# chat('Hi this is Rens!')\n",
+ "# r = chat(\"Spell my name\",prefill=\"Your name is R E\")\n",
+ "# test_eq(contents(r).content.startswith('Your name is R E N S'), True)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "ac64b334",
+ "metadata": {},
+ "source": [
+ "And the entire message is stored in the history, not just the generated part:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "dfbf54ca",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# chat.hist[-1]"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "46716033",
+ "metadata": {},
+ "source": [
+ "### Streaming"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "f02c7ab1",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from time import sleep"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "3fe74496",
"metadata": {},
"outputs": [
{
- "data": {
- "text/markdown": [
- "\n",
- "\n",
- "**gemini/gemini-3-flash-preview:**\n",
- "\n",
- "
"
- ],
- "text/plain": [
- "Markdown(**gemini/gemini-3-flash-preview:**)"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "text/markdown": [
- "\n",
- "\n",
- "🔧 simple_add({\"b\": 3, \"a\": 5})\n",
- "\n",
- "\n",
- "\n",
- "\n",
- "- id: `chatcmpl-xxx`\n",
- "- model: `gemini-3-flash-preview`\n",
- "- finish_reason: `tool_calls`\n",
- "- usage: `Usage(completion_tokens=84, prompt_tokens=85, total_tokens=169, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=66, rejected_prediction_tokens=None, text_tokens=18, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=85, image_tokens=None, video_tokens=None), cache_read_input_tokens=None)`\n",
- "\n",
- " "
- ],
- "text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-flash-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=None, role='assistant', tool_calls=[{'index': 0, 'provider_specific_fields': {'thought_signature': 'Ev0BCvoBAQw51sfwHaoJJ2zAYvG3OjGxnzXwweSDUEiPQ8Fctb+Yq9H/nGaNfcP9TiUmldLfuoOpVqJR6XarYK1JQRU/pctffHmGtm3wUl8Xmvfg1LnyftTobksKcjUPFRlcZwJGdHI3lLGdh67Gh3WgtgY3OoXWsCMI37TaI+8qEtLl+6XLJbahLFlJXnzwCixYRxA2j97jNG6YWQpdYk5ji1Zbd96VTSiwhiC84SQorDWy1o85hLbxVHj6VMG1Aku+SO/sl4wyMZCLCX2ZPZuRYfgR49qaCJk1TI9jA9yQL4WB0oSdZ1Rw2ATPfG2um9AWLnCmRiUE97DIUFaotQ=='}, 'function': {'arguments': '{\"b\": 3, \"a\": 5}', 'name': 'simple_add'}, 'id': 'call_5cdf55b9b81b4279bd3c49b1d343__thought__Ev0BCvoBAQw51sfwHaoJJ2zAYvG3OjGxnzXwweSDUEiPQ8Fctb+Yq9H/nGaNfcP9TiUmldLfuoOpVqJR6XarYK1JQRU/pctffHmGtm3wUl8Xmvfg1LnyftTobksKcjUPFRlcZwJGdHI3lLGdh67Gh3WgtgY3OoXWsCMI37TaI+8qEtLl+6XLJbahLFlJXnzwCixYRxA2j97jNG6YWQpdYk5ji1Zbd96VTSiwhiC84SQorDWy1o85hLbxVHj6VMG1Aku+SO/sl4wyMZCLCX2ZPZuRYfgR49qaCJk1TI9jA9yQL4WB0oSdZ1Rw2ATPfG2um9AWLnCmRiUE97DIUFaotQ==', 'type': 'function'}], function_call=None, images=[], reasoning_content=\"**Calculating the Sum: A Quick Task**\\n\\nAlright, a straightforward query – the user wants the sum of 5 and 3. This is a perfect job for the `simple_add` tool. It's designed specifically for this kind of elementary arithmetic. I need to feed it two arguments: `a` and `b`. `a` will be 5, and `b` will be 3. Simple enough, no need for anything more complex here. Let's get this done.\\n\\n\\n\", thinking_blocks=[{'type': 'thinking', 'thinking': \"**Calculating the Sum: A Quick Task**\\n\\nAlright, a straightforward query – the user wants the sum of 5 and 3. This is a perfect job for the `simple_add` tool. It's designed specifically for this kind of elementary arithmetic. I need to feed it two arguments: `a` and `b`. `a` will be 5, and `b` will be 3. Simple enough, no need for anything more complex here. Let's get this done.\\n\\n\\n\"}], provider_specific_fields={'thought_signatures': ['Ev0BCvoBAQw51sfwHaoJJ2zAYvG3OjGxnzXwweSDUEiPQ8Fctb+Yq9H/nGaNfcP9TiUmldLfuoOpVqJR6XarYK1JQRU/pctffHmGtm3wUl8Xmvfg1LnyftTobksKcjUPFRlcZwJGdHI3lLGdh67Gh3WgtgY3OoXWsCMI37TaI+8qEtLl+6XLJbahLFlJXnzwCixYRxA2j97jNG6YWQpdYk5ji1Zbd96VTSiwhiC84SQorDWy1o85hLbxVHj6VMG1Aku+SO/sl4wyMZCLCX2ZPZuRYfgR49qaCJk1TI9jA9yQL4WB0oSdZ1Rw2ATPfG2um9AWLnCmRiUE97DIUFaotQ==']}))], usage=Usage(completion_tokens=84, prompt_tokens=85, total_tokens=169, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=66, rejected_prediction_tokens=None, text_tokens=18, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=85, image_tokens=None, video_tokens=None), cache_read_input_tokens=None), vertex_ai_grounding_metadata=[], vertex_ai_url_context_metadata=[], vertex_ai_safety_results=[], vertex_ai_citation_metadata=[])"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "text/plain": [
- "{'tool_call_id': 'call_5cdf55b9b81b4279bd3c49b1d343__thought__Ev0BCvoBAQw51sfwHaoJJ2zAYvG3OjGxnzXwweSDUEiPQ8Fctb+Yq9H/',\n",
- " 'role': 'tool',\n",
- " 'name': 'simple_add',\n",
- " 'content': '8'}"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "1, 2, 3, 4, 5."
+ ]
},
{
"data": {
"text/markdown": [
- "5 + 3 is 8.\n",
+ "1, 2, 3, 4, 5.\n",
"\n",
"\n",
"\n",
"- id: `chatcmpl-xxx`\n",
"- model: `gemini-3-flash-preview`\n",
"- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=8, prompt_tokens=289, total_tokens=297, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=None, rejected_prediction_tokens=None, text_tokens=8, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=289, image_tokens=None, video_tokens=None), cache_read_input_tokens=None)`\n",
+ "- usage: `Usage(completion_tokens=42, prompt_tokens=5, total_tokens=47, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=28, rejected_prediction_tokens=None, text_tokens=14, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=5, image_tokens=None, video_tokens=None))`\n",
"\n",
" "
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-flash-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='5 + 3 is 8.', role='assistant', tool_calls=None, function_call=None, images=[], thinking_blocks=[], provider_specific_fields=None))], usage=Usage(completion_tokens=8, prompt_tokens=289, total_tokens=297, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=None, rejected_prediction_tokens=None, text_tokens=8, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=289, image_tokens=None, video_tokens=None), cache_read_input_tokens=None), vertex_ai_grounding_metadata=[], vertex_ai_url_context_metadata=[], vertex_ai_safety_results=[], vertex_ai_citation_metadata=[])"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-flash-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='1, 2, 3, 4, 5.', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'thought_signatures': ['EpYBCpMBAQw51sdH//Bxlc8UYcdMMycFB09R+Avi9L72nCuZTGxZ0hNU7uS6QWmN7R+5uMyaqEomjBqL1mm6lzEKvNGhhh7WMwWjoQ9/y8+kZ/e45ns/sHV+suUR8v2p1R7EE3ZV7Z4lavwWaHf3dSBZQwaZfQSoMT7sPLRTXw39oWlqAvdHY9Jo5v/KlvTXKblFZ4ygUnJ4']}))], usage=Usage(completion_tokens=42, prompt_tokens=5, total_tokens=47, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=28, rejected_prediction_tokens=None, text_tokens=14, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=5, image_tokens=None, video_tokens=None)))"
]
},
"metadata": {},
"output_type": "display_data"
},
{
- "data": {
- "text/markdown": [
- "\n",
- "\n",
- "**claude-sonnet-4-6:**\n",
- "\n",
- "
"
- ],
- "text/plain": [
- "Markdown(**claude-sonnet-4-6:**)"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "1, 2, 3, 4, 5!"
+ ]
},
{
"data": {
"text/markdown": [
- "Sure! Let me calculate that for you!\n",
- "\n",
- "🔧 simple_add({\"a\": 5, \"b\": 3})\n",
- "\n",
+ "1, 2, 3, 4, 5!\n",
"\n",
"\n",
"\n",
"- id: `chatcmpl-xxx`\n",
"- model: `claude-sonnet-4-6`\n",
- "- finish_reason: `tool_calls`\n",
- "- usage: `Usage(completion_tokens=98, prompt_tokens=610, total_tokens=708, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=3, rejected_prediction_tokens=None, text_tokens=95, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=610, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
+ "- finish_reason: `stop`\n",
+ "- usage: `Usage(completion_tokens=18, prompt_tokens=11, total_tokens=29, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=18, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0), cache_creation_input_tokens=0, cache_read_input_tokens=0)`\n",
"\n",
" "
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content='Sure! Let me calculate that for you!', role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=2, caller={'type': 'direct'}, function=Function(arguments='{\"a\": 5, \"b\": 3}', name='simple_add'), id='toolu_01PRYuuvarMbffFgYND4aeGu', type='function')], function_call=None, reasoning_content='Simple addition.', thinking_blocks=[{'type': 'thinking', 'thinking': 'Simple addition.', 'signature': 'Es0BClsIDRgCKkBlwmZMRjyibQQ8AWRB5B+Lv74nqa3VUhZc6QFVjEJCrtYPiJ7f2pG54Q9kV4YS4IR6Maq4kXu/ObtRgkBswbg7MhFjbGF1ZGUtc29ubmV0LTQtNjgAEgzBvG1fa54AO3yik5MaDKuOaXK3MUk3rGHsMiIwTT/4Ol2p0Di3DTpVxOjoMt1knuAH6+2jsMr6LFHSf8vs6nxTooNwuSGfmWHGpobsKiAzrQgo8XY8LoEKlI4WGIDJ3qdVaQWOSJSwk7fXA2003RgB'}], provider_specific_fields={'citations': None, 'thinking_blocks': [{'type': 'thinking', 'thinking': 'Simple addition.', 'signature': 'Es0BClsIDRgCKkBlwmZMRjyibQQ8AWRB5B+Lv74nqa3VUhZc6QFVjEJCrtYPiJ7f2pG54Q9kV4YS4IR6Maq4kXu/ObtRgkBswbg7MhFjbGF1ZGUtc29ubmV0LTQtNjgAEgzBvG1fa54AO3yik5MaDKuOaXK3MUk3rGHsMiIwTT/4Ol2p0Di3DTpVxOjoMt1knuAH6+2jsMr6LFHSf8vs6nxTooNwuSGfmWHGpobsKiAzrQgo8XY8LoEKlI4WGIDJ3qdVaQWOSJSwk7fXA2003RgB'}]}))], usage=Usage(completion_tokens=98, prompt_tokens=610, total_tokens=708, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=3, rejected_prediction_tokens=None, text_tokens=95, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=610, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='1, 2, 3, 4, 5!', role='assistant', tool_calls=None, function_call=None, provider_specific_fields=None))], usage=Usage(completion_tokens=18, prompt_tokens=11, total_tokens=29, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=18, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0), cache_creation_input_tokens=0, cache_read_input_tokens=0))"
]
},
"metadata": {},
"output_type": "display_data"
},
{
- "data": {
- "text/plain": [
- "{'tool_call_id': 'toolu_01PRYuuvarMbffFgYND4aeGu',\n",
- " 'role': 'tool',\n",
- " 'name': 'simple_add',\n",
- " 'content': '8'}"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "1, 2, 3, 4, 5"
+ ]
},
{
"data": {
"text/markdown": [
- "5 + 3 = **8**! Let me know if you have any other calculations! 😊\n",
+ "1, 2, 3, 4, 5\n",
"\n",
"\n",
"\n",
"- id: `chatcmpl-xxx`\n",
- "- model: `claude-sonnet-4-6`\n",
+ "- model: `gpt-5.4`\n",
"- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=27, prompt_tokens=721, total_tokens=748, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=27, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=721, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
+ "- usage: `Usage(completion_tokens=16, prompt_tokens=10, total_tokens=26, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=0, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None))`\n",
"\n",
" "
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='5 + 3 = **8**! Let me know if you have any other calculations! 😊', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=27, prompt_tokens=721, total_tokens=748, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=27, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=721, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "text/markdown": [
- "\n",
- "\n",
- "**openai/gpt-5.4:**\n",
- "\n",
- "
"
- ],
- "text/plain": [
- "Markdown(**openai/gpt-5.4:**)"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gpt-5.4', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='1, 2, 3, 4, 5', role='assistant', tool_calls=None, function_call=None, provider_specific_fields=None))], usage=Usage(completion_tokens=16, prompt_tokens=10, total_tokens=26, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=0, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None)))"
]
},
"metadata": {},
"output_type": "display_data"
- },
+ }
+ ],
+ "source": [
+ "for m in ms[1:]:\n",
+ " chat = Chat(m)\n",
+ " stream_gen = chat(\"Count to 5\", stream=True)\n",
+ " for chunk in stream_gen:\n",
+ " if isinstance(chunk, ModelResponse): display(chunk)\n",
+ " else: print(delta_text(chunk) or '',end='')"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "caaa626c",
+ "metadata": {},
+ "outputs": [
{
"data": {
"text/markdown": [
- "8\n",
+ "1, 2, 3, 4, 5\n",
"\n",
"\n",
"\n",
"- id: `chatcmpl-xxx`\n",
"- model: `gpt-5.4`\n",
"- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=5, prompt_tokens=73, total_tokens=78, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None))`\n",
+ "- usage: `Usage(completion_tokens=16, prompt_tokens=10, total_tokens=26, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=0, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None))`\n",
"\n",
" "
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gpt-5.4', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='8', role='assistant', tool_calls=None, function_call=None, provider_specific_fields=None))], usage=Usage(completion_tokens=5, prompt_tokens=73, total_tokens=78, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None)))"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gpt-5.4', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='1, 2, 3, 4, 5', role='assistant', tool_calls=None, function_call=None, provider_specific_fields=None))], usage=Usage(completion_tokens=16, prompt_tokens=10, total_tokens=26, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=0, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None)))"
]
},
+ "execution_count": null,
"metadata": {},
- "output_type": "display_data"
+ "output_type": "execute_result"
}
],
"source": [
- "for m in ms[1:]:\n",
- " _sparams = litellm.get_model_info(m)['supported_openai_params']\n",
- " if 'reasoning_effort' not in _sparams: continue\n",
- " display(Markdown(f'**{m}:**'))\n",
- " chat = Chat(m, tools=[simple_add])\n",
- " res = chat(\"What's 5 + 3?\",think='l',return_all=True)\n",
- " _display(*res)"
+ "chunk"
]
},
{
"cell_type": "markdown",
- "id": "21fcaf1d",
+ "id": "558e55c0",
"metadata": {},
"source": [
- "### Search"
+ "Lets try prefill with streaming too:"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "6fa66f10",
+ "id": "834c058f",
"metadata": {},
"outputs": [],
"source": [
- "#| export\n",
- "from litellm.litellm_core_utils.llm_cost_calc.tool_call_cost_tracking import StandardBuiltInToolCostTracking\n",
- "from litellm.types.utils import ServerToolUse"
+ "# stream_gen = chat(\"Continue counting to 10\",\"Okay! 6, 7\",stream=True)\n",
+ "# for chunk in stream_gen:\n",
+ "# if isinstance(chunk, ModelResponse): display(chunk)\n",
+ "# else: print(delta_text(chunk) or '',end='')"
]
},
{
- "cell_type": "code",
- "execution_count": null,
- "id": "beed6c09",
+ "cell_type": "markdown",
+ "id": "3b8c3666",
"metadata": {},
- "outputs": [],
"source": [
- "#| export\n",
- "# Workaround for https://github.com/BerriAI/litellm/issues/26153\n",
- "@patch_to(StandardBuiltInToolCostTracking, cls_method=True)\n",
- "def get_cost_for_built_in_tools(cls, model, response_object=None, usage=None, custom_llm_provider=None, standard_built_in_tools_params=None):\n",
- " if usage is not None and isinstance(getattr(usage, 'server_tool_use', None), dict):\n",
- " usage.server_tool_use = ServerToolUse(**usage.server_tool_use)\n",
- " return cls._orig_get_cost_for_built_in_tools(model, response_object=response_object, usage=usage, custom_llm_provider=custom_llm_provider, standard_built_in_tools_params=standard_built_in_tools_params)"
+ "### Tool use"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "bf17377a",
+ "metadata": {},
+ "source": [
+ "Ok now lets test tool use"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "afd9a499",
+ "id": "8daff75a",
"metadata": {},
"outputs": [
{
"data": {
"text/markdown": [
- "\n",
+ "Here are the results from both parallel calculations:\n",
"\n",
- "**gemini/gemini-3-pro-preview:**\n",
+ "| Expression | Result |\n",
+ "|------------|--------|\n",
+ "| 5 + 3 | **8** |\n",
+ "| 4 + 5 | **9** |\n",
"\n",
- "
"
- ],
- "text/plain": [
- "Markdown(**gemini/gemini-3-pro-preview:**)"
- ]
- },
+ "Both additions were performed simultaneously using parallel tool calls, making the process efficient! 🚀\n",
+ "\n",
+ "\n",
+ "\n",
+ "- id: `chatcmpl-xxx`\n",
+ "- model: `claude-sonnet-4-6`\n",
+ "- finish_reason: `stop`\n",
+ "- usage: `Usage(completion_tokens=72, prompt_tokens=825, total_tokens=897, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=72, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
+ "\n",
+ " "
+ ],
+ "text/plain": [
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='Here are the results from both parallel calculations:\\n\\n| Expression | Result |\\n|------------|--------|\\n| 5 + 3 | **8** |\\n| 4 + 5 | **9** |\\n\\nBoth additions were performed simultaneously using parallel tool calls, making the process efficient! 🚀', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=72, prompt_tokens=825, total_tokens=897, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=72, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
+ ]
+ },
+ "execution_count": null,
"metadata": {},
- "output_type": "display_data"
- },
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "m = ms[2]\n",
+ "chat = Chat(m, tools=[simple_add])\n",
+ "chat(\"Calculate 5+3 and 4+5 with parallel tool calls using `simple_add`.\")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "12c993c3",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "def simple_div(\n",
+ " a: int, # first operand\n",
+ " b: int=0 # second operand\n",
+ ") -> int:\n",
+ " \"Divide two numbers\"\n",
+ " return a/b"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "d21710bd",
+ "metadata": {},
+ "outputs": [
{
"data": {
"text/markdown": [
- "Otters are semiaquatic, carnivorous mammals belonging to the weasel family. There are 13 species worldwide, inhabiting both freshwater rivers and marine environments. \n",
+ "Here is **exactly** what I received as the tool result:\n",
"\n",
- "Here is a brief overview of their key characteristics:\n",
- "* **Appearance:** They have long, sleek bodies, webbed feet for swimming, and incredibly dense fur—sea otters actually have the thickest fur of any animal on Earth to insulate them in cold waters. \n",
- "* **Diet:** They primarily eat fish, crabs, clams, and amphibians. \n",
- "* **Behavior:** Otters are highly intelligent and famously playful. Sea otters, for example, sleep floating on their backs and are one of the few mammals that use tools, often using rocks to smash open shellfish on their bellies.\n",
+ "```\n",
+ "Traceback (most recent call last):\n",
+ " File \"/Users/keremturgutlu/aai-ws/toolslm/toolslm/funccall.py\", line 262, in call_func\n",
+ " try: return func(**inps)\n",
+ " ^^^^^^^^^^^^\n",
+ " File \"/var/folders/zl/js35kg3914qc7d8lsdtqsyf00000gn/T/ipykernel_10810/2058224461.py\", line 6, in simple_div\n",
+ " return a/b\n",
+ " ~^~\n",
+ "ZeroDivisionError: division by zero\n",
+ "```\n",
+ "\n",
+ "### Summary of what happened:\n",
+ "- The tool **raised a `ZeroDivisionError`** — a standard Python exception for dividing by zero.\n",
+ "- The error propagated up through the tool's execution framework (`call_func` in `funccall.py`), which caught and returned the traceback as the result rather than crashing silently.\n",
+ "- This confirms that **error handling is working** — the system surfaces the exception details rather than returning a wrong value (like `0` or `infinity`) or failing silently.\n",
"\n",
"\n",
"\n",
"- id: `chatcmpl-xxx`\n",
- "- model: `gemini-3-pro-preview`\n",
+ "- model: `claude-sonnet-4-6`\n",
"- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=462, prompt_tokens=192, total_tokens=654, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=253, rejected_prediction_tokens=None, text_tokens=209, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=192, image_tokens=None, video_tokens=None, web_search_requests=1))`\n",
+ "- usage: `Usage(completion_tokens=281, prompt_tokens=884, total_tokens=1165, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=281, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
"\n",
" "
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-pro-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='Otters are semiaquatic, carnivorous mammals belonging to the weasel family. There are 13 species worldwide, inhabiting both freshwater rivers and marine environments. \\n\\nHere is a brief overview of their key characteristics:\\n* **Appearance:** They have long, sleek bodies, webbed feet for swimming, and incredibly dense fur—sea otters actually have the thickest fur of any animal on Earth to insulate them in cold waters. \\n* **Diet:** They primarily eat fish, crabs, clams, and amphibians. \\n* **Behavior:** Otters are highly intelligent and famously playful. Sea otters, for example, sleep floating on their backs and are one of the few mammals that use tools, often using rocks to smash open shellfish on their bellies.', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'thought_signatures': ['EtYHCtMHCAESzgcBDDnWxxlTP8MhjL0xfAna8F06mzsaYVKOozThexqOWDtggh7RA7GwPoR9U++wQFbdzRMYSmUMDvZBQQsAtXghYQ5k3e8hs9il1YBmG3t3lKG/Wx/oEjxkUuJZ+lnl1CpTHtclLyA9+wyPtcfg1t9tHHto7UlJ+RGoia6Heqy0IFLTwPPu5Olykm+4TxRGHLCH2aR1vKvaNZWH/IMt9Kqd5r/hlH6B0Fy8d/dMoYz/Vw+c8l7k4LIdEy3SavHbApG1JgGn1IFPkdTZuM8K121/3f5pcc1hyqxi4+8e8kDLYXN3h7/VXkXi6ny9cH3w0yd0OSuW3WdhSxbusQAV5Yq7+f1SAxe/SbG5EOuCEjX169SUfyPzslEe7aJ+O2zaX57Ueeev8PxQn7RiSl9IVOlaiRWIuGBJ1cNS7MpxopbFZ8uC8pAr8jqjSc3PbdCfd4JFphV69exHmP6IPQ1GypM7ACadZpoRGL1POO2NggVIbmlAeyg4wEW00DKpgHWZd4okVtuNQVnVWHtebYCqRE5/bJmKVSjVBZmTQ386PRLjMGZelRG0OoyaCEYBm6s/5v7nnmbLZM5Ae8cKv7wfCS9mlULmaE3K5nk0uqnj+6/PWrqt6/eQ23besKJdtWM6s7r/tIV77vxcO32SNfnVU/drWYGx3VsubYXI0n3Y5QTx3SPGF7IgecDZ6U04FMI0ZXxtl+aJ31IUEf6YkmSIoYTRt/gSF2cGBNa7Rp+zIRWjbXt/hXhsc4q7aMbYLjOwgADKETwichyi6WabbUytG0P6CbrvQL06eBuaJDY+NvSEc4D/c+k4OxjoFjG1WHqBEh3te4gn8hx0ZqtBecmdD7scwzfFw2PFeXUs78vNe/GEX4L/JXhQVUIwhX8PRCeHONhKNV5ApsBtOY7700rBC2QpCFbOgpOtIWVfDCiQVJSOhtEe6z843LmqzmGDxMmDoCOkWSa8PfvIYrqp9Pckzw6J24MjiZwSDeRCNV7kwtMxWhV+2T5zNr0rvT5xqPvenqKsAUUFEHloeIYdx5iDM0jgYQ/doa9/NbbOn2iJug5Jvgn4tQK0PPpdhaXoTqugkK2Q435hE7mwS9ifdjjjoVb/d4xw6Ez+CLhVh2ShTN2PKMa7xkzlRBfgECevD7Y1Vis7JGGun3wOvcoztm17nemjgIWycFOjqH+KA7oGmoHbPdP61vJidZCKNSepfyL1nlaqPB9vrfHXdUnJsZpCq48Qgo2jjOxBIAvs3ghVYpdaoAw+GLUTzITr08I0qwUqhsz8LWVoufIGUlouZDQcuQ==', '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', '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'], 'server_side_tool_invocations': [{'tool_type': 'GOOGLE_SEARCH_WEB', 'id': 'x21jhl2n', 'args': {'queries': ['otters']}, 'thought_signature': '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', 'response': {'search_suggestions': '\\n\\n
\\n
\\n \\n \\n \\n \\n \\n
\\n \\n \\n \\n \\n \\n \\n
\\n
\\n
\\n
\n",
+ "\n",
+ "- id: `chatcmpl-xxx`\n",
+ "- model: `claude-sonnet-4-6`\n",
+ "- finish_reason: `stop`\n",
+ "- usage: `Usage(completion_tokens=260, prompt_tokens=996, total_tokens=1256, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=260, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
+ "\n",
+ " "
+ ],
+ "text/plain": [
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='Here\\'s exactly what I saw as the tool results:\\n\\n1. **5 / 3 → `1.6666666666666667`**\\n - ✅ Success! The result is approximately **1.667**.\\n\\n2. **3 / 0 → `ZeroDivisionError: division by zero`**\\n - ❌ Error! The tool returned a Python traceback:\\n ```\\n Traceback (most recent call last):\\n File \".../funccall.py\", line 262, in call_func\\n try: return func(**inps)\\n ^^^^^^^^^^^^\\n File \".../ipykernel_10810/2058224461.py\", line 6, in simple_div\\n return a/b\\n ~^~\\n ZeroDivisionError: division by zero\\n ```\\n\\nThis confirms that the error handling is working as expected — division by zero raises a `ZeroDivisionError` in Python, and the raw traceback is passed back as the tool result rather than crashing the system entirely. The parallel call structure also worked correctly, with both results returned together.', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=260, prompt_tokens=996, total_tokens=1256, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=260, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
+ ]
+ },
+ "execution_count": null,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "#| eval: false\n",
+ "m = ms[2]\n",
+ "chat = Chat(m, tools=[simple_div])\n",
+ "chat(\"Calculate 5/3 and 3/0 with parallel tool calls using `simple_div` (this is a test of our error handling - tell me exactly what you see as the tool result)\")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "dc564417",
+ "metadata": {},
+ "outputs": [
{
"data": {
"text/markdown": [
@@ -6138,26 +6685,19 @@
{
"data": {
"text/markdown": [
- "Otters are highly intelligent, playful, semi-aquatic mammals belonging to the weasel family (**Mustelidae**). There are **13 extant species** found on every continent except Australia and Antarctica.\n",
- "\n",
- "### Key Facts\n",
- "* **Physical Traits:** They have long, streamlined bodies, powerful webbed feet for swimming, and extremely dense, water-repellent fur that keeps them warm. They are the only members of the weasel family that are serious swimmers.\n",
- "* **Habitats:** Otters live in both freshwater (rivers, lakes, and wetlands) and saltwater (coastal marine environments). While river otters spend significant time on land, sea otters live almost exclusively in the water.\n",
- "* **Behavior:** They are famous for their \"playful\" nature—often seen sliding down mud banks or playing with stones. They are carnivorous, primarily eating fish, crustaceans, and mollusks. Some species, like sea otters, use rocks as tools to crack open shells.\n",
- "* **Size Range:** They vary greatly in size, from the **Asian small-clawed otter** (roughly 3–10 lbs) to the **Giant Otter** of South America, which can reach up to 6 feet in length, and the **Sea Otter**, which can weigh up to 100 lbs.\n",
- "* **Conservation:** Many species are currently threatened or endangered due to habitat loss, pollution, and historical hunting for their thick pelts.\n",
+ "To find the sum of 5 and 3, I used the `simple_add` tool with the arguments `a=5` and `b=3`. The tool performed the addition and returned the result, which is **8**.\n",
"\n",
"\n",
"\n",
"- id: `chatcmpl-xxx`\n",
"- model: `gemini-3-flash-preview`\n",
"- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=683, prompt_tokens=76, total_tokens=759, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=380, rejected_prediction_tokens=None, text_tokens=303, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=76, image_tokens=None, video_tokens=None, web_search_requests=1))`\n",
+ "- usage: `Usage(completion_tokens=65, prompt_tokens=239, total_tokens=304, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=16, rejected_prediction_tokens=None, text_tokens=49, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=239, image_tokens=None, video_tokens=None), cache_read_input_tokens=None)`\n",
"\n",
" "
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-flash-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='Otters are highly intelligent, playful, semi-aquatic mammals belonging to the weasel family (**Mustelidae**). There are **13 extant species** found on every continent except Australia and Antarctica.\\n\\n### Key Facts\\n* **Physical Traits:** They have long, streamlined bodies, powerful webbed feet for swimming, and extremely dense, water-repellent fur that keeps them warm. They are the only members of the weasel family that are serious swimmers.\\n* **Habitats:** Otters live in both freshwater (rivers, lakes, and wetlands) and saltwater (coastal marine environments). While river otters spend significant time on land, sea otters live almost exclusively in the water.\\n* **Behavior:** They are famous for their \"playful\" nature—often seen sliding down mud banks or playing with stones. They are carnivorous, primarily eating fish, crustaceans, and mollusks. Some species, like sea otters, use rocks as tools to crack open shells.\\n* **Size Range:** They vary greatly in size, from the **Asian small-clawed otter** (roughly 3–10 lbs) to the **Giant Otter** of South America, which can reach up to 6 feet in length, and the **Sea Otter**, which can weigh up to 100 lbs.\\n* **Conservation:** Many species are currently threatened or endangered due to habitat loss, pollution, and historical hunting for their thick pelts.', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'thought_signatures': ['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', 'Ete1BArTtQQIARLNtQQBDDnWx3soVlYc6MxaosoHy/WGZdw04YhsaQjAMXVEhVWQvsHLJZuVuXf8mWXKOr0Z6p/sDlhDhDM+vDvgjJLr+JvmBmCKRUQCN0GgwFh3GI8o271SStWRyDKfA5KVl+ZS6Ih/PMinHFjSvn19qVpxbsQzXJuKPc1YNfLdNpNe4WUPjLxRUgI5RNLllcTJUgUyzs9dNEreYZqP/O71iX15FpjhTNHVthETXnrvQeyGw7fvc9OIZd5XeZNf/ytCwjqMf071nSLHIqSnwx9h7iJhOHdsRLMVoazeY6z1b6tErcnprUK3/JU4f4BKePCNoLft7KuqJgroJOc5yvI9FWHxZkUqSDbVTe5b+pZqCymZOcJY1cyapG5G7m3imqTSXYLV8kDbLHtlGQziiCP/fOjFbXcgSY3vP9xfAcpy+ZC4iSigxykYSlSYEdN6P9vAamYK8qDGBUUVgS/eXo0cqAlJlkkRq22YoqwAPRSuxL8+6VN4GFBkJw+VAlcAD+Fa5bzBlujWSQVmk26IfDeeZ6Z8Lsum4q3Gdxl2h5mnPJWg7IFnmizgr5BBjyA6BJoFPcyu9AWh+XYNaJCiq1rjHFk3g/+v777Y+t2+T/M6VVKpR6+ShYMlbH2bprxJJsjKklRjUo/qv7famxpHjgZ6z+455IJNtnTuBrALczrfcYeIpvF+LIcAk2ksTeLuMfnbtEVIZcxGle/cMtDcTzsZIxa6wuZKvD5YXDyx7XBw/OHLDT/xtzDFfPi/uAx3oI+Qmg331zh2jKzQPBlkV/JTAcRD3XbJbIvuQxK4dq4xPEeVzGofymA+k/shny7uM92ZDQYd45DjisY8p/deT2/L1YzcWnKZLtZ5JDl5QjtduPmNAN7QDa2HiN5iRpeOQR7FAQmqjw+UxDd9g5qrp9YqiH+PEb1fGAPNfuGxCpKJEkC+DyU0W0Rv9TYdrevrH7GKg593M6jI8Ruii9l4xLIO2S1UIGFjBmH6YLmLtR4xGnDVocCxHfoPvjxFAKOmV0OdAjpbNJy/eX2ZNgiLEd6cVZccivSwHspiYolJMTXzLLu9fYPlDRvT71WjK+W8+b1Bek1qgkqlzULhpE89XzMouaqBMoZ5vswKaHF53gMpI/NwFmA8zCrY1flC0jnBQzvYHzE8UkOjTuMzpQjm5M8t38bMZOB3zRaJzMeEL/R5W9cb+r5p9R0TB4bHoTZETS6vKn/8kQjVxB5dillPt7NjubPznaU8+2zYv8a5qt+NooLesr+KxeJPBVe0vZ6TImQyoMYTeFyXf6mUQNORJV7h5IkS2GuZzYgh/ZzSxqKzWww8mrJ4ivehww6C/Plc1RWVzzf/9PjtiE4wK6mP73zaAVsAMDTZSuz8ulKandcphqK+rmAhg4DNbawssc8fXTelXmyQ5VbA4xhSKmMPlAW61mVVnJPFNRCUnFEvBOBdFDARoRLhBGbQxr+kahZXxdXJEOFeKiQibgl53tWJaCJGDaX5TXCCZP3uG+hpxn8LZFMruYcr+VWhgxbxnh5yonJwhXSkcAewSax5kD6A9fDCBo9P5dTjMMEtaP6KBwof3PEXvOSSlgsKH420T7PfJButuAZjbdgztuLXvTyw3E8nOod1a/zFHM7aJmV8aBG6MZYEwst35Ic5sTnxKP69D/Fw8+8zpYh7+5Xz1+0t5w12LmjDKUa0lt/vaVK4K5QR4Ldp/7yEWO5LTEIzHZ+Wm+qmUt3TNP15ImRQ9ge4SdCUvLO7dgntzp3dMV6FWHACAuOxsLry6adnYIFwrcF7WBHDdV3RoFpD/hiU5gRzAcNGV4CAPEx9MFWsyaHcCf4C6nUH5VHAiXZl7zL2w9TAEwXAtOxUhlNB5RDfe0Pn9j8iFnkPWi2KRWKNm9ZSa0NCqXjtFAo3LcVDODT9PvS43VK+Dnt2MN5Xih3rC+oTfF/A9RMvvhZgWM/8VbwjrXUatuJExrbTT4ywtcf3ADCOuCDmdNGQ/LYZ2t/HADRvxyO4ANfC1NVOlsHVSXXC+kgF9R6lacv1xxReoOWfj8TV0WZlkyZ+Z2/BDufP0WVjngcm1s3YRvKklX3lQw3eq46CySkdq2gPy3iE0PS3XUUUKJh7RwIaZYPPQErMwj31iIupOO9Ul/0HEwYOjyxtPgE36q9Go8l85BJMzwGf7r9GyBD6Kl+9Tk1eFDnj5lT3CEDGRgkVQGm9Npa5xoSFzeUqXmOwSyzCWfKCJspOtZ9K6ATKhdzAolkj0qQNDP3d0xy5z3Zm5ciJlHWKr6XJ/RvjFVuTqcEuQC3AO3kDMV7mxMxB3afOu85XJhZJ14NeHChoH73VF5hPqRLevrTvkQwjGbr/LcbaGA79VrMnt+DMhGa73nClrGOT8l3KsyaApz/J47hiofws0MK4tTSYIb6+12cB58ISg+R3KDSbE3zSC0hqyo1REAiBYvL9IEkHqXUfR+hFB1aszeCasrKSprZeAs+W2fY5W3XSdLtlsiV8w46dw2WjPpVnwe8hU0yDgH7QfzL6IuJ0Ynffq/Sts+jxG3m+1JI/UY8TtmZZ+JsUKTDAIAM1Flka9amo/h0OU7zzXUDoIV52SZmRhpn8JUR6uERX5d3zDREcx6SLJS2hqAVAb6VKFtB6BihhF+/K+P0WOlDqtUiiXWrdWEcqZ1uLgq++duTIeBSXoQ+5DUlugMeaeeBhYg+knO8tXrKR1QOEk4qqen1lAE/k7IllWOruiM58XpnXEed2by3LJLREfECCtDhwJl+V3+WvVrUSL+oyrvgI+wtbijXopqDQNOZzVr/bMHIbmn5pBRKrqYRCm2I+RVVeS8EGf5I+WvmQJnH5VkTDXrQJOd0XU9TpSG7nyR17g8AI5/wJEkhimF/kd5SGX9ZEb5FpsL8udVKCY0qYDZ1TKKXA/tJ7a0VQ2mYSo/vBqtSStWkeZvmZvKImht7dBVpFBaKZXi6iKHJL7iUiaeDJsbkNPiJegPBjdnBNAzHRKqs59rBXIAKxWfcDAisTa19BUmKJq8pRxPLexZL99ICoAxf3iYvc7voTMnkmA0SkhAGQl6yVf70ZdH0r9voWhe19bmDghecUzX+DCwhkC5aWfoTlKTEV/61b9KNMXg3cHIOzBny3D9OrlJhfeCsX3qhtT73xiFnps3+DNQYC9TnPaccKahvzI784osEk65+0vM20fvVzQ2b9VrBzhquJKWBcjL7nqa54Trdc1wZ2hmVHcxCZWuTFtZvLxPuPZ4hzLi8qeg1EE/woZJWDGMxWTVh+5KV2KcjuyItoGTh3Jm8mx/gNxmyOCdGJ7Ho7hbWSNkPYsFn0sDv1T7ATRoW4miNdx0SuOYR1PWNT3yWEq6dc3/0cIlzhWYfPgG/cvluoon7h0TA1iC3UAXZzmJVajWYUXKkICLYuoys9NFQp56fLkIU1gEFfoFRNtgYni5DNuTIjh/wxT+5oXrQSlekauKWu79O3NsUaXRBt7YW+e0KmHzZQAAmxZlt/oo06AXJpczgDVLZR8eMZeERgho3QmMUaiF2M6j5rJ9wxYNk3j1DvPS0pKnYYNK8prEOO8Vu3f/awbifNwVUVwud/vqKdY5V4EBA0J+o+87cDa2zFrvU+ToQXdMYhf2MgZ+z0pi9Q6tm7JlfoqqVbmbmpmUk2RjA0u5lI/5LKh6kUv4y8ioRrGV8jg39rmjDihp/fpK04pMjKIetP5xedzUk5LMmgIyroQgMhvEqPBpU/I5U/KymUXcqFKMIzmwFWAzi8GM0DOOHwldORXNei0h/0P+Pa6ZSHlAnIav91hJokcT/ubtkrO/YmzRvL5vHxNJj5OJsMInEzwjwZNrMItRedPQvlTgNVE2GTxkC22iE9lfGnS7zxBY+nRi5+zCF/HvmA0g9Cw/FdPUOt+2mA7qycQoNMPAoifjdy7A8tuedWQXUEqHPL9ZTkNohJwdpoI7dxWmmGg1bM8dgxOpHIwVSCjV8E85k3MqLmfJi41rdLgGBGo4WPlrL+HBa2gdIPGrzQOBc6hu9BAQWAawg02jyhfIefHStZpVQU0Q1mqhZKwA5u7/zvDWRG86tZMUSfMF/JdEUcmCuLj1+0/u6Me4upXzh2hLWLYnpJyiUjVEUCpcusAYDXvu7lSmsL/fK6x8HRf/xZSSmtyGtQwq29/j9fyAZTLU6KnsZY1s+sjeWwzGuolkrtVXeMr0CEZo2uk/mF1b/PLbHoxfFKDHVhBXe8r01R+Uz7jKv5M5X1bGpAqk3ZIDM0ca6wnJouJSHj2af5nAMjprGh2s81ncY3Xis67PF8DtZZ/2HXNVDK1zXTGokMNSqx+1QBN3ywI6aVfLe85CDq0XBoCmzPBtWrPhSTE7gLGXj9vUZzAYDKadenbvSn9X4D70bIMB/JBexBGy1Xa6K++T7l8H7o7BLmNKNPn74y/0twpyIegg3plEF5lAlu3+asoXvDjTpQ9jkoKEDLc7cRcXTLZXLmc+eo4GX1U5Vh+IEYzW/ZgVtVpuspl20JX5/C1jV4PzFp8MAk9Rgl28N9ao97lvXJffbg/4Gd0nj2InOaSvas7VSHdpCIWvWg5bQ5jO6rXTQCl406VnO2oHdjtx5L+KIL4ClypWCAYzhe9GnBVdthWE0R6bz76oI4i/dF4u8EwaZDB92URKV/13/9Caz1+8o0Q/xMJXM2v4l8O+7bZR2fby4g+qfiA5s+0lzSK6vlkDPMKfhnjrSEgy9UQChFWTUpkz1Px4zPlyu+Y1Ioaia8LbgqnAyxD2vg1DDBpEf3vbCsKdOo/EshVeR0m4rxZpUk+T5dT2RMcDMRuPCeA7j/nqduV1cPTcqbqT1kwy0wxy5uWCWV1kPtSPNRX1vjINs/E0ca8C0HiEDjNuR1BBzPCEWSt3cxc+UzqpuKjq7vbGqlCSiCazTp65TVGvM/A4NmZiZRpwOtbOqYeKHcaLQ6KYSD65CBzUwZT5B2tXHjS3XR0P6CQWcobciuF9c+B2gBm1Yv/gYxGhqp6claolP2wZyziysai+o5NEtfNYqW/aqjmX+bi+gwh0FRT8Lkj3p/Uf2fGM8en4UF6HgKRrHqoeBnpMyX8GdoYra61aqykpe4PHNvJ9NwkW4nZu5wfBsguQ8kvZ0T+1JMDTEz96Xvb7Nz6+betDHDECRU+L0WN0OCe/MK3ZsolfjAASmf41yTbejVOxhDLOu+2QYqbATHj8swjJ6KHkS5ZMRZ05LUiSF38HOi6iA1YSpJTOsY1PYTzyxD4xYNjWuww9l99oByzMaHOOU4Fptv9QhfNXkuk+81FRtsH4SlCvUw/N4Rr/3mL5DJY7gdf2nYY+6cGYX2l9YScnkcDKzsvwGfzyxaKqozeypCLCsIwWO0uVOpDAKJkLyR4yi6gaGzn3p0AaOl65Vn/dWSdT38pVfs2lqj9A2WnxdOyEHxm27o8UwucJNfH2XQ1CnK/DBdWlxAw3RTsWsVz7x6psuZRYX7ZbaIpnUyD3jt/3TzXGi90HQ0ak9Dc3GuwAT5AoIOho7ykpPvfZNMCKhcy50H/cZkq481ln8NjE5M/F70IF//02KG4MI/TQhO7rildWTazUBtvHha6XmM3NgFjqz15ujl8Otb9hxBGBGw7bP3aoPbt5DcZzIR83+dg8SpsTbOq9ycRr5fjMCdJXReZ1L9p+qulLPOeHCdizeHbHhG7T4pFwREFu5RuZtKZMyUUPcMtrzK7hSweqNNQkmcyhYGqAjW4lA5/xZlquNNk46gcdPAoOzzUnQRXmzEwFrTdRYCUJHJmL89Va3sJKq1/H4K5T+jCBfWukS03cLv/bEZjy9uNL5PgJS3fg+jaoACO0R4/y937Y77AVYqy6oslH5FMKnvOic1/FdCwhMX6W1w+lM2HaJ5QeNeXRK2eLGp5GYKfGyRWrPWj8NwN4j14rhr1/e38jNi2EOKFqYO78kcHQZ/kfl/iNFMDlUwtr9t6wV7GciW/4hSEoLJ0JenqufYlu3eYcINy4mrJMT/RhOfbFDPQPBcoQ1Dy44o4Aud9WXjjmZpz/OHlw7q8vOiY3Coa/3oksXMF10Tw7zU6DXnMTI5lH5JOmDT00iTCAQoLmjQ7uOtrqbKRYzQIRtLTVFP0MWul5xjYQOGB+lW1OLDuSuCbTsoy+2C9V8Sm9fdn6O/Z1kRMCHOMBRSOLhv5mDnQN0OSCsLYwtnKvFCI4pw13RkIBG4xAayHZJaeZ+E1x4+hOlD+tklRGUcduvRAzO3Narh6Fx2eh8LDTwjKpkftfUTHLlDN65+4Sbjif/Qc6qupt0ACvO6oIR+JyVlVkZVqp8vaOSfRlTRnughng5LNq+MqvFpnhVJ5Y+JQuDNMLONXz4CVZt1fZERh+SZy0KOVaoiWFAfwxfo5K/pWrglXsRoYUZBNOO+PRQZlKkou0igsxtXEfP1a7Q9v0T432sspYt6bImwyKIE/p2Xm7hEU3Iavx+xn2bizXwcS4f53Fm5ZvErNfvCyI8cSbG2lTAYewcPUsRdQvARWfjaLMXvQdK2qzmEEWGLUGbNhyaosNvu0xyz4KXPDaULXoYkSABQtFE0SYNpgC8zRvwjwaaHOhZg0wh/YtgVBSafDOEwUCjSuksr/M47nFqTOIoO8KaQgQLF99SsTcoLLD5NezWRyRXlOASWBdj2GSsiSYk22xXAGcfBleQGNNz+nP8Bv4RN2PCJa7X3Te+KnYWShwITcZ8Rsb2bRXsdtMh7laFwJkwjETHX8np6/iUCEYwgRalN4KJQ2RVj8dosy+iua4SLx+lll4NrtC6fbivnwRl3u/WxlbYgnhAYLXB/Eds2yiEjg1mNYnm/nqMHgjfW0Y7FIfEya0aX+2oF8e7mTz91AY3PLHSXjQJId1Mslz9NjDfMYqP4dcyfwAycC0u+856j0S4tTmEflRTddr0FwMjw1q09c+LWBHP9cqhSaFfXSlI0ho05KYJzFEp4V5sInAz0gfDb3gX/FjvjzqNHvlQENgg/tUCauCJHecThkc0ACLJB9NOKZ80NnHiSXMVSQDQbLR9g62UmoWQq43ujHxhuxs2QggmBrkbhNYvH9/KWYc2exZKtckqOMbEdHrE6RoiZ256oLfyFNDy7IJx6AbTh1nIXKVwd/K5OncxF7ipjVrPnMkAdS8/Ocl1DRY3gorzPvDn11mjwvJJj4yiLyaErxmiALtpHLNk7dypSYgxMZiql4joBwAbDglvbl0TYSNoBG1fjqbLQN5pzWv7hiK3KrmOjBnXQbWssldx0sLu7dbyqbziSU+//R3ZdnIU/bSnUNcAkE/G19FvGBj9xdsRxoid117vZsU4LJYkkTM2PLIon3ZGlZ/tw7oGqxojKUnx3OG/Zw35YiULH5dM3AIDsnL0q668+U8aOtNcinCH9JBsFUrupuSXNTmkZPn8OJO+DJOMOoJg2qKkwHq5LvyZIF32yWWVHPok+9VQPObSNZ0nE2tfdkp2K5lXLmO2d5ZaPDZiSUGTGd4VIJH53aOLadU3jcXaHhKaSl+jIiycXik2MXA8JiCRULjBAXWGEwcjEOdjuty72H8aHZRU+mW2kJS9jraVTbBl1Zxfl6Njgv/B/cZI2hQFdws3fSHbgjvnND+U93sxvPxQigHwW1y/kGzwnbHQhjenCZBU0kBu6Z4S/D47s8mDMlB9SYBoSXATFT3SBX2eNS3y8mjqLx8r329oI8B9dJZTSrsevnleGXF8GcAqFStv/7JjYeV8iYKFmqr0Xe4zfDxMS2VB+F5EX4UouHgXiIL+pvHHtUI7vYr6lKuDqbbMNzaQs8/nNOgEmpMWHxnTes8RxrGG43YWQ0d2Has7Lsc8NHcCnSY7fTFb+3jhvnu3neS15yf91BIexTFF0EC82K2W14HHOCncVhGv95PWMFxFCOrdMT/bE14tdiFftQ2CVQFs0xBSMgc7EwsqqD39MFH+l4ou/pvg7mNSUak8hnMQ66vZSDYyWNOo/6fSbZU2Zil8OFTSInL7UywBABAl6jC5zqb7Ki4kCWGfW2gOuyAqjES/cDitY+Yt7xxi70Myqx4JMJMjHPmMjP08XyC5bGw95j5YTnf8h8ZrKLZy8RMLcWMi+BZ8gy0+LCbLRAcKHDBkm/X9Ezdm/I05ll/4ecgyusUVoeVrBKhh0xxiIs3SLaclxrOr7WBGJ6VhQDOiVhwJyxfQvEYTAkZS/KEZgadv1YkI8RXCMKGyeXqY3arNTyOGhL+9mBTuVdwh/bWJLMdx2yE45nLT2B7OUZe/XNe7Ku1BrN3vES1IJyhSAGIrCCZukr2Lg4RTiE4IdbBe97vtjjT5CA42fyeBPc7XX38KOZxuHRC0h5y6MUBQaCRY7mcs4bvMmaH1Yoo4hsHEMEIpHFmW7KMvOl/SN0G7Q3ooverfrDPSj8auqQxiDQr9qDbrxq3aNsAdjYw35tbwPCjbE44fwfoCsBwtntyux1Gu5PBwQsKMEmWyMtfzhJ3UezR2EKOThpyw6hpzSBXlOg0zx5UwjdbhJFonF4qBUorEJt6Titc7gIKoJrKoR2r0n3Npy2tTCP/FI5h0mx4Hfa1vmhaayfXiNb8GD6BOv8UVwRtMcXwcHuoyBXArwtF/3ShsB89P8AsE/kzareVtLMNK4OUS7xTHwiq11t1uxbdDVYTUf03Dg+1Oae8htSPFrH8q5tZM12sZo7eBv7olzjMIgNZzers43YAJw097FF8Uwe1xCsahmyLfpGaRqzqWuoSzmiL/0N0D2t1m5Dp6H5C81smhs7o3ekaPykcBHuFNfp+HzaaDapK5qd+XcoWQ871VEaDsQEEtdjfN9C2t/na17pts+cVBc53wXozEvrrzaRr0tGD0wAfM0v4V1JCmAfbnlxJ6b6tDNDrXW1WQvJBrSK1woQy24G1XipGZozptzvC6e6rXLvjoUl6qKLzDLQnemlxi3Xs0cPbMClIrnBEZz1ungTVgendyRJj1DTKcDcFnvzfMrzpSf0+KCmI9ZYqH1fR/nmzuJgdfRvwywUcRWrDm3/rfzSig7XZ0TQCzzbDZwVk0SzR3bpPw/yvyio4LswH+92qQz6w444FGsxzrE56g7emtLF/tXRVWA7RygflqIVJB+oTeava4/2zm9a/J9+ZrhvYc/BCLDKsunq9NBJR8Ap4v1MA1NKEd/Adr5g1YH8Xofr1XDKZctFIhz5Yshm2I0SlVnSjci4A/wc/qcvozUdhax0rB7CVsg97BXFtzLdYoTCzh932tNhzkuTfBpULI9UiebwesXjoqj8eIAYRy7SlwzjBmhwS7+/n2KTIjFMPsLjTnRE9yX3/+jW9jYQPGsNU/dn6BvtgOESKtpw0Zt94PVTmzIAmPKtVm5d0AKFNIwZUB4iq2T154ycdjHlnhwSQTe6DDQ89WOqqGA4fJPYOKiKeCSyPp2yGbxd7GgcbIWJ1y9WoiEaMmbnEVAQxo9uOHGA/O1OrzRYdMmi7i9n5szUDcSKAurRvyKm+ZGnP6o4tgUfFwl90snmw5qvBsMdfT+2iNwn7roKm6MvmVb3Wt0Q0x4UvpDSLhtA/i/Sizzd9HFfcq2LHRrodWLvDgFz2TDIIYTWLlD6+pjOihNbOX4hZCO7hB6Wf7gIp6MxAEt+CaHV8OZE+zAyrujwrXman8zqcNVZwMx5qdaRPqh1HL1UNw5OFiATfO91nDUEONaCyWH8DIERX/YoQJ7AQdhOQMwiNYRuMczyl4jIQYfrDokV7qg0jeJeTC4rI2NVkUkDG/qfasqBgsE754vWNhdgAJLlv8QGHxc8LPnoiETZZzcND3Rkaz29q5QSjU4b2aI+rvOR5dsQrvYxy1iWBz7j9mRGAsWs9rOLb7ux08EYfSuWHmVzOCGWqLLpshQnCXpsxzXiwrml4Ahv2Z68EUnl6iNSNtCsKPrnZyw7BAHxF7BmZSo4DPuTP3JOJS+hXTpRKImiybtO3RIVWrUYocEP6I+GP52k5XYLS+I0yg01tQy8KgOpyhDg0hSWzIVqJVRMJ+Ihnq+VTmLFX4If5gQmymAmT2nPqFbB3AOvUlDjZ/9RSuUVkJVGwWP1HTPPmZEf8uYj8xh/p3R8HAm3BBSwFolmErsk0vaxVSzjdul1a//IhPvEla0KHKsJMHzODGd4daY6vkTXE2m/eNNFKJEFOLyr0HT19ixWgQN7WooYuRH1WW8+VFZCD/G/Eqx9YoLSrzgIbLlcBy2j4tPoAIMsnPf/+D42jivgszClaM1NSavbFm/NB2nU8e4R0us/wkB5zVzjdZKCI6T5HejB8coHU8TnSwPXLLowbbThKPBJ9Uetstsix0CmwrpgUPPI45z5+oNvyPtnfFDo8/Z7p666Xu5QdpG4+Qxee8NAzopiHJEuLSX/+5z9Vwcq5thE3xBfGwJ5xyNZQXel4r2mA22LVaLTygfgjCrHW39vPpDCcPUG08Z8fZP/z3V5GUTd9NwN/x0OnOlF+isyU7FhrvRhHohBXVpIrv8he1p8hXm9oDQjiu6Vs6mynsT6qehCAIzlMngzHQMOXts/b/ne5OvY0fdWnK8N1l0gFBTFesCh69+bZlxXbS0QNRcpTfVE9y5BJuKwgvMUBXV6rZgi7MXh7vykGteY9AxtqshsOO+0hCZfXP9cU2a0A1cQYe21SvXjwNCIy885Bc5XFBHj6/z6JVCZz0zgWhMnW9A/EIMC2cnkWkEuMZeWkRWmi9KjjUhPI/RYplNlKcWYuxYbHIH4N1gS8bDGT28t2NeElyFyXqVrjOo4yxROlxrJiVJWWn26ZjReTMchnUlYLhuWnVypl9rKEhvRsolPzowMFTFjKUDYsz3XpaQufs72caompqIPGvuLUejR8OjPnvAelh99XHLUOqTpXH5LCB7Y3SPlRogjUs2zEydmNURdGrppH3wKEMIdSDx+P1yeuNP/MebbmMx+gIuiEe4zs+NeHT0J7/kDwJAISrexo4cr8meodaCmATJDTDCoXGqW5IQ/m7olz2lpCHdoetn9vSWVxBWyGiTOhK6Pz/vrVZDkdHBXiuYeDkdvkEINkb1OxgDjukMoyKOyAcASlBsepj87+uqrmEjsW33iTfUVb2v+HCiCmaiZwoFUxSok1JsBaseWMLnQ1nKDVk7EgnMKq2rnm6JpLMcmKAuMavuy4hc2O5w3urFduQM+vJ38RvRn57m6oT9xW0D5ya95zZpqTSRnEP4BxKKZGQoSuvInvEiZmg0P/Nnz4hkIDPOGZqYjyXvqNrYFQjK/x1BNSMKEq1ekvN2gEC1wntJq/GDvpcgENTQyH6zy7Eu/vfqZ+EZxWjuStn0kfMmO0uYAFDRvxclVnPRCn1OIxzsln3NeaDcLX42tOIhGf8cfpqlNDWr40Rn8iUlynnoMavr98EeSHW+IpkZG6+SHDUe8sKj8JJIpje64Nl7mdcU4RkhA0cstaJvrtq9nWqk2d9MOO4WISrB8bBDGcsy/Co7jfF9Hq+g4weyu7Wj9Ia/U4FzK4ainX+NQO20ltONI21SVxYh88vl6AUkaXIOdeDqDWQ4fs1r+uPPPe4LlzZPdsxgERufz7zJq5jhfTQYa7k6P9QpHxt4RqRCx7IxcER2NSy5+ZKPkHg1zwy/UQgJjYjGVVTZtu0wGE5LLu0apwNSA++TEf+fl7MdR0UNYDDz3g56eMwvdRVer+ImH5A0G4HMK7UMw+fvSfKXhKNsuFA7v6OwBAei+qXkZfXpJusHf1f1XtAfj0ewuv3Wf7/LIsHWvHLuAwL6htyhrpt1t+KG+Jqpjvv8EVWcjmi2asV/mqkv0Lh/KgefvJz+O0faGQ/D+3sS77I6W3tULWaV4AZt+bIKiUE12dWZoGUU/H/WgXvk1jUJMLYoIVfx3HzONj648Y+wgDlmbUWKK4aPbF0z2OJ7Lq+UmO9/4gqgWTC/qdq3Rd0UZcM6ZIuMI5ws9MNCSaYNSNfE/OAFFSBYcl6h27ETWDEH1qYGv9gQ96A/qtmXQmaTYqNMJ1uxdbGLdur/1PDhM9Kh7zeh/zRcOrfNTSyxxAS3vy4fyJxxVO4k/UcSLQq9ip3Ur4j++6zzUZ6RlcDML3k8PUreqn9e+hXQNBWxLd1mnaGUoVfUtR5ZyRXHXzDkawkmugvKlZnONvUTAQMxGKUm3vlEM4+21+I/D0OSHAuoZjOefk4rOeDTHCc6VY8MbLqkdrNpdEs/jTEO/QeBFmUpvIa87n6dMV+hIefHsF7nYkyKtDu3epnm+KL86+3FFtAjn9Y3FWztXhsJXxCLRS9DexEFNCFEzWbISx9e0kaUDBcSwa5h9gnGDpd+s4tYjjXDlcTL6jOrsn40chaTPiSkxDenZkI0GgHqVqEVPB0jnOgfOAnVjtqZBEN6Zaqio9Lm1TSEPxQJCpykRNEbcxIfOw8HJ04WeCOANB31akjwRjWjFB2/4lWRxS3DDF7s5SnEQGgFiDP1QCINxtU6Pow5VLXdzo6ZR98hh4jAZ15rX8MWUA8Lv4jNxRZ7Xs1l8kILwnhbRRkO5m2J7g3mcQZE9U3AMUWkbJEBSOpi0vkZH9bFVTx/Ev6rb2E/Q0xxbwmLrGy/VClauB8AnpOy15gAgW8qBTBdnPv5Wcbn9IRHK6HYRtHhGrBqSuv2Kh0v8tzptiQzzXEBT6yVzlF5LR1ZIAPrwewHdrc977Pc0JIBbGp7BZ3qKH/PIRRUayTKcKESpXL/8+/4cN1/GOx3I6RROoiYZ0O2Xu1Rdyw4Ph3sXES8956pkzkAyaR+5eAFwLyKIt6Hgyr7Faal2nV4GmfGZux0MfF+t56wAKWI/+1V6CmE2SiAj2Iby1ZkdMd1LM6ozVpGUT0pbMlkVj0+H0B0AWSewVnL4ZJsEzo+n6DZt94mFFnS1ULshMzd1gSgZn/FVkOr2aM6+ZwqJ2LglW6OCUYEPE7R+GNyOlnC6MGf9n0C12VnHrKpqK1/ih49tesoXFUrM92piejfFjamj3k452DEOou8nkvfCgqCMpHPF6fJsMm3p9dpLB3u0EXa6n1kxY60oregspy1i/VM0yoTThwJ9xCUYCAOmd4rSXhvUHflGyD71IxXK4tA+0qOsFq5QASJbGT5QaNdQhi697KIQuLybNTaqOL6FGb3hNAqFXgk3jS7PRdOss3Q3q/gR+PT68PGcSvy45EeBMj4dyCIYAIwqMSvAFRt+VuOw3GuWkVsQPGwu3s92GGkL5iO/Z66Lr8w+NH3BfbvRjr/CTRSIUAyWJSeaMntSV/ndrpZDxuVdkAB7u5blMbtFBYrSmvSpRiRJp3bush1DRc3aabr+zUt7jeVPmQkXZHxQJlRVCNV63EiyuQWsYCFGOz0OkGcn8SR8c72+gJSi+6zq1Xfn+B28RG3an996049kJQUbzdPGVku4LvaATwa9EqUp7feGTqx4tfePv/vY47sUAM4pl0+8Z4W56W94LEwnbmvKG9kLYW6OVp7cvmJsROsB4cJ1LGbDNRTG5XlAqPPC1+mC5NSEo22XOTxou4zoMCWOuGSCTPpXXGv+85S8Lfg1atoJWXGp8eeaEp8RtgkvuIXOrHrznroMnD2nNeWSNC3ySWckxDT2+Pr/OBiSnpPjhC/rTQv0wUxiXYwrf4yy+Rc9/NK4fDxuMBmIE7RNg/X1FiDH7ClYTIHx82TrSRDlgKkIgB+mC90eOXw6rHwJmwyGqJst1YnuQ0qI0ggJDmRFpGWUXDtXiUNiDqjiE7P7LUJV+WB64xvt0hc7aX0PBd44w1c2llBKzse6C3XWL3zPTr1PS5e5UkY7n8b4piBv/O1A0O3iuEqd6Vc8hhvSzA0kA8aXmuGoAv9pY4csjbbnWZ2ntcG7YI0URHymAoGRpt+DGhIJ/PhRG5QuYvruCouis18nt+ugqGln9gU+3HcJTat0r8L5afMCLmb5zMHkczBSa8YEt+qcOp8/ImZ0lTBhkOvCu9X5/Pol10srmJ1K5Yo6k3hqUu+2PDVj9rBlRH73MXL6K/hGZGICRYCHXPdo98zOOhbm272DIqP3r2enJ2XKf9s9JtSzIt6o1WvQsSi0gKgLHOifIukANP1pHXiD3Gzlh4zkk7IpC1Ml5aTy6eDD8rCR+mdFd8X7TqQu6NrpAEnKhvjDiqjHC/bOFu2Smf+Kl5JMH4YRTJkFiQK8emdSqGKLIeKfqQfRznGK9zrKy0zCqhjLD2XFVX5VKUQyUqt1ogebrUHUr9kQAVCppBh+uC5x7FvFj+KjJyGPzBh66g9v430c44H3SQ9u1Oyxs54gSklIAKm8cE4uICAu3utg1xsXDr2WaCS8Pm3OvCM/cyyhXgWRJqeuZx9e2sdl1CiFByMsbFX9sz9ImQS0pA9mcrwr/OeyVddpXU0YW8O4xdNFzY6fsAmul3KCaPfr7vnX0iUbwfHpD8nwqxtXgcp8l0ydMzKipqiTase3fILdfRhcYnFHODVbhDiy5dZD2YY1a3uvHPuA5aEGamRePcOTyffq/sR+i/WZ9goY5pzGMN6EfA5ojvR6vf89Y425kXLeOGfBKgjnpoqHl1eWcoVcznrq5+UU1ZoQm7J/NcaH1NUxz3wV/Zye9h11evnOILnuLI3035QPgBEmZxhXEGU1UKmnHTMQz3bzGrUztkkeUaQmXFLWjBZhVLlXuMexcdVr31kuSRdMCLuhr08S/0NOUX0xF2KDnidM3pMzkz6G0gty4DuaDE9wkdf5ACbOuepktr+3sHHrP65b991i5r/cmDROUhFfkKFqlLHxBZFlfA63WGCC1Lhl7x+4rQlMZYjYnfA+2km7bwnlSvNWV0pp8a7e8F5CMyzZ/MnVlMIKO1rTNGW7c8WGH3M0VoR2qndhmwHEeIeHk1/QHYrv+MM831jAWckmLrGfa5mfpKtMejjs6btn8IgiWMvVZ+D73pjQZ6zRruzDdBB9AFDLDiFeF+xkgp5RrB2W35yzUWtC9Q9mv8uxNT0Rp+2XcE9Gl10ykp8VavQYY0jbVn6NshJleVhjHHjvaHXOrDhkZqoPNOVPsUxo4sGaomGqAmVS/bxW+Jn6a2ipGMdspTeAKid1UhbAMDFEkcLVUq1v/3ZBjjvgQ0ZAIChcr2y/DS6I/yQTWCRtJvqVs7QPvP/Mj7BUxCc3JkzOoFYqYbfikqLIE+gDLNvSlcf6i4r8ZZA55kgr/LAEy369FDznRAFOVXr9HpOSB2BVFnPxsDMWu/RMPySaSG8SZV+fkKXJD9MS/SMmSITzcvijwWZy66BqF0GMz+ltxfGxey++0gFopbCGLx8dLmhcO64d6qYgNky6mHBHX5WHz1gZk/xGW/A878jNHjV7vozMQPG2OqqNjFAPBHMUE6u82fPTVbrY6SQHhb62kxChmyrqyuVTsEo4bkq8n/u6oVX1Ijq/bIl8fEwyxyIRcGrXlLpo3IyFz9mSaryygcivOE8PBjubm1F6IuGaANIjSDlezlXh09rlNyDUTTD353yAg3uw/jaUoiAUcT8HRBvzDaKKNrTyeg5aGJtsAyMIVkJWbUtgXu4Q6f1ZI55Vhaz71b1oqDK5jI+s6LaKqTV2LJMNReP/svhuVVPN5VsnND0ngGVHVPiG7Gx3TkdT9SdnfBRwhMANGO0rKIdj8qXUPMScDlXAaPqjwS7c7HU9rwpHOX2LAOJJiXUqNhC9ph+MnwUhTdhR9XgLJXAMBF3qtD1N6HFKo+SRWq8OKR04+wNyQLdsQg/ZChG1mfCL4ADtP/L89ZXrK2ClVaO23l9/TTZ5v7YldJ+RUyoYGoeYMSEuWu/vPn4gdaiXZM0zY3GkQ3vNKV64w5/c75mq/YRR/Jzm9baO9aS9ei71mRZ8Q83TUFQN7kImzHFdOKozG+F/VLUNhxrDnrfRTOkHhcS8CvvY3Alq3CQomy8g46wfPDKu//mwS1WMS8aXsB8bAhZwEWd6eSCldXRBGEcD6HiJ3W8xTdlVH9evrRVTRg+pYmZ3VdbCsgr2XGDE3g/m7GUKPTcMQgPCxLy4oEszez5oDJDkzEClIiV7T2HgS0uGFlMKOf5u4kCCm7JkjEsbpzROU/zf9wsTCjMsreqwEMvf+A6EC5hwY7DxelZS8bSgySK5XAbn/uYgrkzRuRA0gq56qKAQgkRWfkCVm1ZQSt06VhtxIB9/rEkVoPA0pb50Y5aPG2YhUxcqnNTY29pEHZ00ijcLFdHqj6xYmINs5+1FA4Q27pZqrRxh+kJrR/RNwRYn7bkRDklmaClHLGLzl8xONYxhmBdmLUE4YQHJV5Cm33f1wzVO62JzlnVdr5txU52rZPqa8y3YMvw35QMq0LjElmSLtfPAwYVnphSERBc7lIDJ75fQYpz5YiPHenxQw9IfkayyvoeKu7hwHUyRuCAh5iZISc6Rf+EiqZ2t/lqF6g4c+ZLX64zdBGLJnLnRWeQGIz9eoT7rZE9f1Lg8MkaG0fdgO5WB6iUqglmMku/jreHpi7lnEHDGEEVxOIHMYb9sdhBsxEwMNrEJnQ3yjpWxqUdhE1SCLk79zx+ctKXmxmDMzdWpSYzikxCHukwvPr6cmDIXrswCedRhgVvMTgitw0tXBpNGwOMJ3d/Flvomu/UsfQ153uzJHAeGcDGaIO1aldrNywF4VBMaO+TZgKhAtYu9os8jnan1MDMCsEz3YhZSUECe1ghmCr6J1wAfSK5U53ALTCHJ3ILeoQfEUOKPzzWiTfZ00bQ+jDkVN1V2zAA/8BLg/vRF6zqJRkCw0UWJXxbUelUQCFjp1V/nHREjzkOKKmrkFH6KgFKjXvXzVcTGwGM1DGu2NBv/ikcyU1ZgzSOKejss+gbWFy3pmU2u9myYr5ci4J5/D1LE4Dv1yM46lAl2ytmE2AqUHn2MHtIhGQk72o/Rm2LSffm+xZIEIfnjfm3kpkcxtiDfBNmkVfx3aFLlI3V2iSUqz/id1AZS/eWuF5IZJq5YzXXkKtUbEA3EQ19FKRWG2TFPoxOHJvXomTJY9Yr3WJgf3lSt/+x9c2g/no2wMIMo28AHSEhjIPB02WJhWGunYiydVmD6kEofPkzJ6YYG2PyMq/IpFQrksEtHcj8MFlUXy+c0PUFBaW4CP7VN0EwTkidpRrS2IypxyhQLHmf4frDlbdG8XNowXuXV+z2+g1JRQxcpUe+54lXqwoJCeRYo4pcV7aNBzwrTEZlipqikkn+XXTiQ3oJ6Us7rzjN1tILTO5zFWZPCEn9lMc9IpPrBaQscpTRB1DWEBe8FAx1xLJWF2l+5Yww6BTBmJUkSL/QDvrTFRrk+NPsH7qi+U5dAmWHcZIoJuynCbW5zz0mzVNbwcB/cEKjEdprBrNTW5sxc07orzccIBzbHX+zqGXSbmReOJvkLyYFEgyo9J3R4UkAJqqWcKdM+G0ioIqumKShs89RwGIQGp+/r8nE9iGVQkuYropAyjj2W+3N7tsoKhhVtfEMRb1G4mTSZSka+GAIyRvHAKxcyibFfSzPEtXFsLjSdfGNe3CnhFnwpLHLJ0w5dM3Phd8cLCojMRpmQsaQncgW/3AJfbY9cntZg5YVj8OzPuq3eE5Bdl3Y1piN4pNfh/dugk+NB5xwTPvhBbBAwKX62RMtt9nVSAHJ91dp3A+VdBcTtyR39Yw4iNN20wEAV5db/g441EineLY2zL/SlMTsp8gDpcLuWsyfsFrJ8nwpio//rUyiNUF+OJn6/iHytTETj9wBv3VEAQX2ltyAN/rxRPJ+TYMy9y3BVMkpFPzaY9y0WAL+zxc5MOeKJCgMUpjj7mIlq+NTX1qXhDTd2QdMgGRUVEALInlT2vFW2NNRVJ5mFtnkLEhmGJ2830MYmZ8hZTV3w3miUQT9fNgPxnlii+aZDpqT4FCGGkZpc2ervCQgywhnu5OmWl/yGS8zylPuVy3402HnRX3NAabTa6jGVHz9OTBSYVkf7djnAVuiBS8kGYt9+b5uDf7fsJsZqdylI9rM4RhVazykzyIARN73qUX7iQHyxrfvgsIcsxFJQbc2qkNKOmZxshPebgyHtJb4O4FZqelVS9+UrklmPAfCvcwtWHAxHSRzR506YisWmvFTvZzSDU1dppFRF8ggbTyptwviTkl+5P9KQhU6t0zqTTLJc3FS/Y31QDzx/zr+fEtK0opcCSIRBpWMr2hv8OnAt2rFcFCXhc73oDrexUZ/WW0z+e+JjLx65qcJojTGuqU0oBn9n99WXzwFm3/Nh9gnSORgMnGAKegU+TSe6h886/Pk8JvCzzUvT5TxbAaebNlyFmJSr6d7sTFRh84mkU+p5yKIQrKrV4Z2fL2xVQc8J7xMbX0Am/TQkiZPr8qq21YxxwNJCcDoEOP3KONzWMs+s0w8FAcZnvi0K0+ax8QYrkaC/SaPhx9MWJ2Fp/cBqhhH5JGT82Vceg+C8CgEob3Sux5wQREUjrSwbMLpsyipdMkuE2tXZyVBHPe1TeCeNZu3Ic73zfQfLBb7NrXYYKWbo3QOyv2ocSMgKPJ1sEow6WtSDJ3JDrml6x3Eq32euZ7b2eZWrHk2tF+lb+7KZ0RWfvu7P7P8WJZG7yKrbQC3+4R+u9EodCO7JJgiiC+e/Xln69B6uKcM4/DjdmK74tP0/TePHiGGrHPHcDN/vDvLfql2fkdoyXFXmrnRHOFSL5EwVF1C+GhufpbazGZTeI9zgHmjMfOY7jjKPaLNN5asHQhSpKbLFqNi3eegYaD8G431F14kEmTyYkC60cgEWsZlKS1XgwVL89o+IcjakSc4oh1rC1ckQbz0kDtQPFslTONtvtsRELPkXyTte11LpqMzisJNjouYcW+Har5RLG2iPvPDd4UXDqGnxwsVKd3Xa+m29FV813qH26e2bgUeaMkhB4Pq9heQfMhkTjhEiy166MYMyIVfM0L5lDHrcUtrichBsNjhYRLDjAzSDWlLpwWc/iZDTiQ+fbYbXmizYQokm/4qD1MPPS3D4r55bP+ydm8PqTEro+4tHuweLEzmWOgG+BC2eI7oAigKJWSOtTgFpbjrG67bm6H/Khl8hpOfmQbqq1quAa8DKCYHJ+DlPThlbTOHcUSj+17PJA3OgnO3papwtsuHvmKQldCGRTkls0C7Ccu+VulkyhjfZJJNM6doIowmdeL1CkqBoFUYZG3fpA6GgOT4sgO8lCD74MVMPiE9B+5ItLs9zhiudqmBY5//g/YTv+B38ZLFimZ/wuKwpe0xmNDum5W331oaAKqhmpSklzAReqrGQfee9j2x6TN/o5uMpHQXf4dNvE4r76R5P74i8k5Fk6/uZmGz9IRTOYFZIAYQ58k+ElcLJcZI8AMjFK5f8tv8ojt6XBPiJ+2W3q3RfVz3e4qvcHtMjY7ei8RaTmZkfsORxt/aoH3hj8Nygcl+d+KNxPYzCppIKVFqJvZCWNO+83+xK5YAEGiV4slGZ3FVAUT8tDIIPYEl8kK4+EePMne3Yi+zfxJQQzXibNhLpDXphX6nMD3YHLqlUDBDv+ZPbf0RYVQC1LzNF+VErRYTTt89Y22o9kcK1arsNX3K0bpkapcCZJ8s95FJL7v+R6BI8dgUGcGsLx2qvZ9EeLOiNQxU59FePaZbh1HJpjdhOhle9jZ0cvo+KeBTfGzRIA1nYF2Bzz81AhCSWAuxcjAUPpr1mE5bN5dCUDm3Sa67EfzH/+mUyXTLixb4kCdGQP6/ZMIECilGQ0hKpqHAINpYJUCIeh0t20RDfreqGX2jfrpocTHcYUA1nVr12DzKcI3h/1mEco7K3f4FezWOe/5149kbq504P4uzeXQ3ezFFk++ziVUk6ZWxL+00to5gIGfB+g094KTJhDo6qTdLLNjDV4plhH1onsVbNRAZ+77nQlO5k2jWmF0RYRfrs7tUWfC9YBbtFt+8vpsEq3+X8QCrK1PREJnIJeciaBoKvjVd1iHR5WgUq/X1/ytDDhJoOw7kRsl95aJNKP53d6F27ydjv9FezT5RsgqHWRXQpbIxC11CyUvCcdeFV0VEBqg9DXmnDoX2/zaXNgv5OTYBSJZloDmqD5zPKdnRE/bZwyFHZbXIPw3Ql84wFTgxfDTCkmuTPz9Kg6KRPglbj9IbfdUGW0h0nm5UJb0rMEJFLOmeOPrivjWob/WODj9JCC7RMPg4gG5S7G/CDxWJ2/O4hePnXoUDYXixHaAZuc0OyZWyKXA3wlETgXPfCLyk9cr70V63Og0qVZDEqqj1UAMGq7FZs+cQx2cRqeaTh1qTO41e8DJwRCRQ/UvR/e8HSB+v/Yo30vPDVMkNxp+9QQ4hwiMY1kkbh0tibQTOVxSADsajpx66EG89N3vYvvQc8qubPrHaN/Ud4kKGOKeC3VAjVA5eLX6fw3m/4me6yVzK+kA529Ie6CLwNO3/ho5vq8aoZDa0YJlFRYIXvhGyhw9iJrDpm6t/SuxCZP2iD+tlvX1Q6nRHVlyOXKOE590sjsqYdknbF+8NfJF+gAfcQYpfy9N6ZB3WYZ/5OVKC0MoDBpZUVd9qXlqXpdZmlagFQrgc2uZ31g5SeuqZnbtC+3OXYLggRTY8XRkSu1lhsalLKWSdEjbrfiFyRCwZfZs1xq7bGgWdq17FQKR/m/iN9qgDmzuvCB+9uIK7l8DGaJRS0fkd7gSS6r6AN9SJK/LKex5cu/69Q5kF+4njT39CUscJZF/tEGi3cBqmVkK5HShA37qoCBKQvjfJL2+VWStPRoPifWzZCbtcu7jOOfn3PSmPGfwRv1oVhHBaHY3TH/Af79DHe2t4y6xak1JlDXGOokLoCmdqBDjduqJMmXm0JSOIGTC70yPD+FVJBJo0ZTnP4bkZPIsM4Zq2v4UuUF9c/smZ8hWViqAbwWxLTJiisgTKF5lu1BgSsEWOrU5fmz4QXUmW8lut2JPJ31hgzU3kgOkWDWkDF1A0Q2VNlfhFP5ZSdegZAX57dwvqr+AeLDiMJsf/kzX+Oifm/D1AGSItRyDGNUOPaoFuo7LTRrR+2iQba+rrhqniUtoNqdmCVK2SambTcqxeBvHRXz56BMwlTXPvTqVbwNmlb2375zaW93+CorUOZBag8boxXGFjXoBWNWHnzqOCdSS8vrh0DoUxURBVnO0iq1ZnQw/bdBfCipZBZ9fWiTf7rJgPx8GgCjYllOyvlpI9/fclIiviPqVhc8mVD9nM4CRLXVJ5p1daOj6HWbaxKX024Nmp+ID9uE8ejMKwjoO0KStaX8Zp1dQnn+KqOD2E4+9J7Xoxnm6vqZwTVZOHaTqXsjbT8h1QZIgUPUMiRWJxBn7cj4HAU+Em73FLftyE+ozXFexkNU1aj7k4K8xCgJ9qH7rZjtZZyLqMuFP2ZUEfYSI94eXQpECAbVuiApY1RZgmpQ6GmsvOHYpMXPMfcLGeyo5sCZvvf5o3gTLVInFpFoMpsm435/GsuGhJzF54oLVIzX1qezk80fa6KInFlo4fVy3XP5A+Sryul/zdCD9vq0IkZh+kMxq9DIufLuTDRmtYytMPCKuwnH9tlEaaYWcig9LSAlLV03Fveq4h718bi/pCzqJKXB0MdTrYLjdpCcP9zJ/cveUGlCdjrUOwbU3XYDbtdaXPXTRQs8TIm4Dwmh0SpZ6cN3rdFExGtxzszZtmGD848IeX9JER//+Hn+Z8K4Y+YWEvAHBVm9Eqd1CswNClAs3//296BAAhLgeEVq8C1B7vUhtJTLoWXmdOIu5nzhSBzQbXjVNLtQPYNcCrErO/yz+n0LfWALAh5ke/tWYFS59ZtxWKdymB2g5dkAEyPlqsNyeIJWn8hxA/1kk9smGyl+8K0TP34fA+1FydVTfehVAumDGVvyDsD6ANV5t+D19Y/eOmSqOjTg0qrnu8ZNcRZ+jB1iYRdxsfbn6hh5OdKud3ThZvn0a9KBV/hmKGpalhCq1hGbdOHXb3ARyfqPUybjHkfajsscQGhqRbqw4KJOTim+iuHUZd9XTuKjwSCcfMJvCRlr65OYjBcMVKdezMwMXpjA7V983M4T/LrrBpd9OC3o8uIpmYU0b3nkdvP1e78cYxydViCMjr0xxpflSuKGZHcYtttueghD7nV9fs4pJlWxUH95oJfwoynDVmfDHYOKpdNVWNe9x30yrb5tnq5B3duzTC799Es5KMyElQ8k3+3CdKwekejYgWoT4cpLWLt36G+sAmtKepg35gG1CsSQMeraCQFq65z/MgT2XaFayniq7+VZUagNYivu/mfx2LCxSl3Yl7bLXWqxpqLmqAqLFAxPsZA3TB/DPGJagea0KmB4LxtRfrg5tfj4ddc1w3eUQEmTMAv2unwgEhgkkteRpy+1usflcUm0PLWt2ZcBJErTCqxGAMTVK9KGDTfe+mstWXxKcJi0UBpvqzIMEu3Z+b5E3qDuvjPDyl1IuDNzlibrs/qDivGViQh/9QXkL0Z8wLT7EDAplQx14hOFLrJsMeQncX3822evbZ9L0Znk2oo0IlKS9HAxJz4urxh1lOAHdWgMsDfNud1IIi44vMGoy12/bGtJ8ZtJVwMAhOvIZ0vM2zok78QdemuYhRK2cm+1l5mzKVHzNMXvGYLDWnLYkSk+LzwglOO0S1recirp7TQbxw1Km8GBap2JKsT9K1AcuYwD4VKf5aELuzxEZQFP8caSWfyNFrGDLB1BI6iTc9faJ7LvqrdLN8L7tzA7R4Y++FHX9yjxhpHfx6nLS3aWpr21fT32/8iYCFGkV41DMuluZyMpRX95H0/AIpLCLiHddyDK2BZ3i0bHghPtiImo3GqyB1KBujiLB0nKOJ4Iac4/6TND0fKfkNe0JX5lW7w8xS5/iYYytnQtsPZxIQyvBy3IWgY7Z2g8He5vi9wMk1/q/Ds8srnhwwvQrvZW/MPMlzjhxjtlPO6cW/Z0bldeyPtkJJuQapObFNeVRpm6w1tipzZfVAjuZvKckEuKl/C0sAi/J0dZ5ItpJ7CMT2TBCYq+/BpaXqXz5BW8t5U+m2SFij7OxS9WlyCFkRvXdf7w6ACNNqkg4aNPQcI+Aq4W7ahnOiR9/pUL57GYeEiSGBiqqKai2ZNCDzigwZ8OlXN/s6OfK02krtxx3PL63UEXetJsUEeEam0zkos9r50AQG8DsU31cJay7OlIx6I+B/PqbFuuy1VW+0WqGZgJn4NBfavz9L+KrQX83Q950LkXwYEQEJ0XoWtF68GUohbHW8moT2vjmMiIhWJmWU8Nfl4o1D6wnggpl+E1YR/MamKX1JMGu+BhIcqhkfFH2qE+Bh9qT7DjPsig+W7NU5BcF9OJehm3JmC78EWMHIMJNcHK71y+Y0qxPUKLKjeWeqx+vS8n06xWB9akE+gvZVv4bMIyC1bf4oM0gZy6o74h9/O0dTfrRdTffqPdqWFCdtV8TarildpCHEjrtAM5jjQ4x3rjKRjIdc/t790tC2lcV832UTuXrOpatRFVfaySMs0WygJCTKkojmARKvCqSKBbpflIUPBakjkjnbq1EQRnjLVOR9BvMIyuBcydd9NducUpUa+1AWZRHoAChk8gSHw/HRqiKErPqj7fPD3R2hHK6BOJOXLmmQUCLdcovd1LHR9MMdoLVOl3xy+xB5jz1NY00k10Od2VWFBIjeYGLAmkMQg2I/8Rg/ptcAubQ+oO3glOV1P+w1qHgNd0hnRxVH6xv2hxb7rYN7+4qYY6KiC/MLD1SI622jdSvoLL3g308kIHhoNdcyrxcnb0+iUXNIl3wNGAv7UBxHOFJI6veCbmE6OPC/RHY4VvWCrEvmjQjvGxhydZoWkIPrJm1bRdd220i8AUx+S6FgpLQA6m3litGkLYThZAey32A5WbpiZx/YQDGZRxi4VPogGZ9t0wE9XRKT9DB0nk5uonX1FZB/lNKDCxY5nWwp8qWzOnDLfZpboHegty6eyr+0V3U6p5XgHRo/7k100WtTo+dHixvR8smMMlDaKbtGV82SApQzn/05yfRHitBmEyTGTH1Ia6Ml+ofTuuJR1Coj7JTZel+Ok/C88Z4EP/0qzU//jRJgx1YsnHSiBskDBTHqtN+tLy9kX/xaO39/kO7Pwlcy/33z/in+AfgKKr29Ohi61eJeIHsuot6oeSrNGqEIKxVEOoPOqtWdatMi2w3QZb84oGeBK/2n6/YTJX80lZSZsB2cdPAa832D83pjGNUyUla2zKiZyBIQF6+eXBOiPJO9AMUJE0ezovpW0XtjFg3tvDLjVRP+2aQLgF4FdnWM/A/Ef4sdd8m7HLwQYckYX2/2ZVINqRjXV0JhPGYLAidP/WV9Ue+kx9KcQ/LE2WeIl6F1t6+mxsNGcrzedxc3jtcxFRR8A/1K5PxboB61Sg+XkwB5HB4XW5NFZsCkvQwM19R+xkGHGGFN+75XEUMsGU6vJVi+e9ffhJGew5IAhiWwexnCzDFlFl4JwMUivTP6huLFAEDQ3z1ri0zy2j3f62H99KkiW3pzoUzVGXX37GeUSBw6GVUwIp0FzxgBEMJPADAc5Rh32JhseLG/c0O2MJPTUImIEYI0wNnLRba+Y6VnxyX09QlUhA2yYPlX0ktOz0/hXz5vlfqD4Z+kjfmp1+7PYmQqGcru7H+402pLexNxJFA1ezDXxbU0oYWP/eDu1M3IN+HO7lBRFwJC0yLN4zTr4Eu244Cd/aYGm5WfvMT0cGF5S8JIBI5aOeokxXTsVSn+Rn43zBbIe6wggYrbENWCAFkDxX0L0J/8At6xKLgFV5Wu2yXDwFaE2He1/c0ZDobn+Uw56n96U3jYNUdoKOsohy+7DrpsmBQmbbTB198IjNEtQyfWDY0tU/6DM+5lEeLe21W2zAsyWESQDnDquA90C4Q8FeQQ+WpRzwN1H6sjdZyCHVh6sQpVJoLLoR/g1o/LwuK7hAkdXnSPj2MPN0n3JgxhHXExz0udcusmyic2h9SCXh+4rfzOtNPJqf6Gmf/7LNlvA1BglOE1P52ReHfmgJcPQ6kT3JW/NXrM8nNOTjNOxCPyTVnUWzTNHQ5ZV5VbaALqKItGoalfGaiE5EYPzEYrdOnz4XbZFxxC92MdRyOxDHDoVrOz84KcWSVHSYsbr11jholqD2BG3Nt+BcJQmT8wRJSPWVBd00H2CICOogOw4TGrW/VGhBEvHbeLohp1Plk9fi2APJug+SlrtAwI7I3VTLZDggnxP3Dy9JijMB92w4Igwgssmx1T6l5MOf5HjAGkMOMtHFYWHbpuuoh9vyjw5q360f/h0dpCTphbs/Ul8Ki/bEEN+lsMDJ/YSvhg03orKIfO7wrQTm3v2XkJGfFbalEpeSf/LgdC2wj4h8DvBwmd8WCrWTLbwnU3WLwcC6uW0N71F8CYaoM/fXmrkSGTtcZBzA0GMmMPqMd1U11CMLtbJLHQk4z2sYdbuTpavJdODwlZgvsFloUEmZh7/olS88Kd9IuVG+/Q6X7vOZNbwxW03xofq8j9e4Mw+6rpuaWS8ZcR2m9auA67S9mwtZW7bK11yp/1GKw0U/dphv/SNb11kfHialTneOCFGlEZHMpQH8VwUOecetkuWW6x4IXX2kK94EO7BX4PIbXrLrTOGU3HoSUZlrXBoi03oxWPohMsPAUH7Um95WstGcFWdt22mAMbDjjjzY5Rd61R3qjYNJ5UEWUtiyW3LW+hpVO79XdATf26Do/dmqKr68Mq3ebrXyK4sWHJR/cfHdYzR36BKJ9J+jZcXNnnXqo0Z7/oWYTDO6MbDm8bBC794EUvzxu3DeR23VJQLdUxxRDMsIM19KucQ/7t12EgMegjQCqg0CJrNhdFCzY8S4MiJuILqAZSIO0Qh1ArZLfOu2MVWqTiQw8+BQaCZNVwkGhwzVvy2zHceIddnIcTfBYFC8ZC8K7sc9f273g3iPwCF4bULJWe6QnFivUmWIdb0QU0xufX4f26VZpSQfqapTQ+fGPLn9ci8H56I3HjYI5OAlC6Z/Vnw2vXy6QyS/WO11be9zeqrfo/F01CqDDAgqNKrspC5bTmGdgtZnrV/NqL7oJ0kQu4cMZzcTQ3wPO1w3lUGJp7zfMdBL+tqVXlPuIx4bbwWacJ7xDskyIhZR0BS9SbKVoOVg4f/+t5GLrwcoceogHkGv8Fhr4ITfdssiFwtqr3Db79CXXhWiqmqaxvXm5Dcu1a7hEuzsTvzhJEEhwBPC/qogliQpoIOBJM4e3Im8k8NpmZQfYm6WezCkpqTzJltRMjWJCiZ+bHo1CGa+RjpE9Fbg2ElWqsgQtXXDrTllEriuNo23Wo6euzg5layERyUOitW4lEYh6wpeFfEsBdHHx15a8OZOTU4dnZw2+pXM1tx5pZU1thEtMVbtsGXsx77ccup6xSUvGhDuitbiS9u+nu+vHfoLZw/+Myf+p7BGoTy8feTgT8toanQ+uR06sGkKHcRD05my1S9Haw2DUpGobspUC9w5Vg5XTd2M7iiCTJf2mkQDOk01JIaJdXVs82fWUBw8OCfSfbBxIQVyY5fIwtmdSKgPijLh/kZPtQ5QZvxbEPqvJjFmyu0yoQW3TS9BRyIYfib5emTKmAgfvgD0SMF1DFQZ7hWmj0ZkQsZv4PGWP8RwcvCGwnVfVoCv6nZzxSYPzBp+jrNagYfBAGiPkpE6egboZfGwNY5ha4UZcUNp2iN2JksPdS2Wlqe2Jpf5NgyefeEE4l/vIumC7j1VgrUzuD4nsg08c49c1yHEAHEv3EMCQSoTBJxhKHipHp+SdGoJpg0B2MEBGie1KF36RDJwHFdJzVk1uzAD6RYLsAmStyEjQww/wvsUtPa55iTYL11eEix9PdlPK1Q2A6DUHSTT7SBlcl9bm1WpOjhetnxwJ1qmCEgEMc693i9K1W1W/upYENlqG42ZbA6e8EPoPJC3bBFDpgxLpBe8VfT6/yBROV8FTPVagzSbuzDqTgPIuYesUOduun0x+LOcVMtbvg6NVne9jbSy8L65B43Whi+Ak5JYKyHpSGynGFNl9lnD+zb2hCKkuKCFQyNC8u1ZIWOgFZVG5zETs0VJpfPliC/WAgjhrUcfZzd8lh0nXsQ8JZJRPBPcXUJRuIIkGSUcJ+YsxJEMGO5L5wVf5LdWOmB/qu8hEjegwAKgAmGOIq+60QkKkTU0DoWtzczaxBvdYLDzI8Tdxqy2ypQZ1Bs4GTBWvyBw2EeGwziI2K1qWQiqqwRBaT3amPNZygX98BWmGfo82kRE3avjMk4Rj58l7ETZFU0lPGLqHut3HJKxAccQdg8aldhyIuW7l8iuYn1sjIk9+9jONH43cvqiL23TH8eyIxp3ojvlmmnFoaY0xjNnx0nmV7rdGqmha5w143sohNzqQYajYOM9ydJt0aEoIBENk3RG8k6OUl5eujexYEbNsqPsRjE970UvggEzI0gdbpVUCtfoU7Hv/INJF+SHLwHieryHCu4gv1VY0/x4yXrzw8fScL+DENnImR5sPDGpMz6FgjJCgi/6nCE9BOf3U6wcGlxdCOXdWk6WQ6DElbl0SuqEMRWBZCYwqMCyG62vxcq1c8aKrk5ATGfVzvhy5STyeMHRgr+LCMKblMJf6Cvv+fJkZVSRA3NjhvwqdcPZDHr9v+Zx8eRKsUwyEciaSdcag3U248U23HyqJ7n+Vy9PXg7QCTLU+jVgjCw+795XE3LZR+7o7awTxS66Z1gnkofp4c98BiQbmS8A12YCk29tJhPFngnMvmIAwUFuGzi/FCSKJ+2mu5wMkb03/irk0QDg5bpniCTW299J4MoF0VmlBtkOADcFZzVrYnO/upJl6Wm7gWa1jMiaM7zzcuW0BgiBt62vOquzM3puGs+bcst8y2Pt9dmkZhC3AkI122P1C6O1nNdlnKoucPpxTojkAeAtw1CB02rt7AcRxpOKckyRKQ6chCkWb7Ns2xajCAZfvV4KA22SSlBXIy+vk/jtd2h4ajHtLh3H7LhvKmII5ZIluRFacpZsRHOI6cVpaC7ByxQgw67aihGlvYsLhHvyQDOuxP+to/rIdM9XFRkozJzqcErgZWrrxvXBnFd9Kj1xI2AgqtYsc1vmZGgw8KcXErihDhk2Dv2LOTRQYVi1/c+1t6LfzguocFDXtK3xbRW0EV4Fd0Fr3WuNnpl/G5jdzD6JY4taFUKgzeq53ZL9g/ZIay/9Fkl0u4QIUMvgp0SwgfcFvchrC4Y7hKFfaqMqt/8K6D9sqV7WXxGAYQd5vZ0RDrEVla/9mYmg5G7JbQYWz9Mkes0ja0h0BDscgduy8Z3TFEfBw/TdTG97wxBvN7ajV8n1649I7EM5o6RV9RUIkPHt+pE2BdtemGOAVmYdGUEVsnXJkqFj1f95rJpdKAyJW66Sb0qITKQkf0TwpZjQzCt9VgzrKFG4IYuBUVGJL4BW+0WOXIauOv6UvKf99ffB5J4ZbiigLAYWSmRjhfNWWgBxW2purZEYMrOl/dN2VJsmizVBkBlYo+VQi7hE134KpeyjKuZEuhlrpbZCAIj8hTTLi0d2kpLYyVD5Sz/jl0AEb3Kd1NSu0L5BlYk45F1kG/XnpksGg3Tm37xVQz1llTnOwa02VrBMf1ragmk5twdx4WXRhHuf+LYeqBgQrW6GfGBERGPl/PcUlHVv/sGA+iy99NWooYAqj6eo/PWz00kc9RmEN5PTIPht3tWqCtaxAS0nFp1xpFRC3HXsGY2p18+e7TtEm18EGGx+GQ94847YgwwDOsmNYIotKyTNBbjo2gWVBWFId3fiK8gbXASER1IJuJb9NBeOuvEBzdqfNcjDWO8dhg63YfaCRQvUlds4VKDoEBuLoQ+jQM4nW27sARsmg8qPDX31aILFLO9SZySN1kvDwpn/TSCPn3AueICeAN32fYqi/fwQ3ex12snlCWGP4vpKGLvB/GW5V2KllNNx2vGAR26WHwIVBpGjB49vTy09AcmLZdnCfqAmAmZMaGz+xA6/rWsmuewPAj7Se6Xf6ULx6W8iKIFctUty0rsX2OZVOsZlV1i9uYkMnpYANctaRa/3J/9dmmcJYPfb6X2ahk0V5edwTpH4y90ciJdi0rMfDTiYYenmNY0ugaX72mC0wkgexRO1W4K8hXbHmk+QszbYxnG7OO1qIn2o9bBm5HVnEs/KDA2GthzFae5sLo0iUbknVz033oepwzZplnBCiDrRQRRnlXYPHtboesIXWmffiCtHKWEuUjTL5ACzhR2Poh1tf7gBHgz4bregKX6bQxXlV7k2Qt2Lwwu/SqmR2iO/VRINd5g6pFnT9oQCKhBfmSi4og4WbLK3KRy9Go62IXlSt4lpzbbgHuCbkZ4XJt1dwYMMpkJa7X0zlzpNH40CRXkauJo/fBR7TYuIKKkCVnlbWmGX5ZUTi7SxJJzE9DKhBk4IkRyzMPfwHsIfYydRU+HwJqlY3OBICr9ohH81gxP1Q6qwzz4LAPJYuiC5+n119mNKQI7GGAVFM77iNwrVjZNfigKeUWLycbbOvo7rBY73UllQtPRBZRNaPeu/P9lT1oC/C/Qxq5tlisWtJDArYicILeS+Ujb9rbtiWPaWsgmyY/CoG88P/4Diu5s0+TWgVd2qLU304XRzrPddrgF4VIOr+M0MXij73vjPcgRBSqJx6yNxzNV20Wj2bp5qfsPNEc6JdEMwyJDuExCAOee9L4CxuUtzjg6brj37ljrpyRmOoemhOvk/ojQPh/puUtrNmfjrh/8A7ycrrlDGjahprJ8sYTHUcQzlzcOucLwS5ugUfSLpPoYTR35GiYY6d4fhzwk1rj1EzAi512Xb3Uo5l5dkrgQyb2Q3y2+1AkjvW5WgkTgXXpvLrrE0zE9Y5uQwa01ajq6RffyiXaOpB36IdyfhRwJ98wQ2XirIa+ceqaeZGlvDGCn2unjPX9JDK4bzY8Odn9cZwgHRU7dnfPjg1asPwxC/kNUV2r7SN/BxVA/bTdt3o4oU6XCtdUzK/73W5N1YpwIGN+c3RE7lv/1fP4n/0/EdacT1kJIXy/y3BwTtT76OX+xHDKuPtnquVvFeQm+Wk8PXS3dx+SC+k9yWLh8HzxR4vqz3PowxUTrT4zO7Hb0SGh4eHxC1lWb2iPxdmuTVfi4jPTK2DG0XyDSgPABhd9LtGMbY/0ObGDS36qGhLVAuXh7JAmb9A9jmgMlYpZr4mh6o/MT9D80zdKFn85yJQhSVb8BHhX2F+W/Rg7dEtk6+p7+kBvHuAnrKsRfrJyuNWJFbCxk4cAVajOteb3MZrDJUzvjKItGcmTCMGdf9v6MZNxQINqoviP0MbCVNdjE1ggTol0nF4d9b3gVc4v4ElBFHDYdavtFysv1/yDIJlhSlWjekmPm808JqfwYoXHhfNiVStmlo11EqQMgcSHKFq0WU9MOAeyrwXwcJdsgw+rh/b1iPh33qsRHpHm1YE/S7jmuRgfEDER2CcWJUP+vxZOBxNNzLeA0UgFXfE57HaCagO3J3i8hJ6GLj3uaxlrDttDUHnbgqV+JOoUpt+E5HvmKR/AwTj5Q84Ek0kwSnvwwZPqJhxh+IUBrmrSwrBRT1AvJ6kLCwPgfsPFBgrpeaiqNrj/vMy4jAkwOZPzAc1fU8h30LScKNW/ImhZkyzA+8qPRqMlIeZfOmDmbG1vBuY53tm1F7J7c/Fd4UkicSNiqN5xkemjX+nNMj5jEJhFtoSETYgFPN5u4Sz25ioBwDn73+AKSi986moJP36240kwYuyHSVG1qVH1tIINRSlaf3KwkDAt/xo3CIm52ywc2C/PbUySgBtkCFXNwDO5mghEvEjzyz4KIgvRDIjROyh0t8B0lqJCYW8ok5NzXF8q+lgcENMdg/F2c7OBhxEqynFaFAbZ+9V955oukqnbUBgHk7ArKfNTHfQM62S0TaRtJeB8MfxGPdrezaYrMQeedLxKzHnVPLVSeCvmeRLdWVNnCMmKeyVQSrPzhw9/1E0JQg90+taygIr8n+zqCGcIEeY9bi1nyYqiHQ+i43ST9efeTj9/BXF6Aker/3kDyFu9sLJPWixcryZOJ7LWnkkG7gWjAJ82C4v4ZJbKXf+U2GpxqxBSskOqSQWesMc6NrX7xriyO0UaOteW97n9B8ShvITnqJoVJhgIENbWNct77iNCydXHzjh2gZhUak6HFhe5+0f1F6q8H/HD8H6p/kSU0sSFooelxZNaFkDP7oYrwtIuWGmjpziKJ7PMVGl5s5vtUSGoGV/6WKHRanDQ5aKsWFIzgziZ9ZOGlMTvnknyXf6sCRY3kgq+pPhMni6zrUiw2yVd6ui9bXHhUlvmvUuuynmWrVtXc/YRhWExPZcHOuNrzqZNvdU/G3CfPXGdaIrgFuO2r/+3lFlsh1lCdyV4qbGuCCY/Yp/5h6UWM8IF0k4vVIM+wwus395xRlpJT/94yfusVctyy+UvIBcV2iSgjKz0JbCisWydtclKMFZKiYCxN7mWcAblzqbQWsb9ebuvngGic+j4FEHB+BP7iMgQovYx1xdQFVGPkfdlbO581Ikap/BPInXET4+KsLyn9TGnk3IZq7mqi/MQr3jlel+BTUDMmuEmONHeppNL5boW2KOaRbcMA59NgUXLVNhix5IKbpX99TOJhsPcaW/mv3L2N/C0kx2hTbRo6OrbonTduY7ai7a+oAmqwgAmsVmv5iU1wWl9iME2DwfxN7V4J2HSF2o0v8hm+YW7volttwlKHyYhcoM2C2JXIfq2g/AYI/W9HANHeK+Ck9kELu9GNE5cU0kirPpufxy9Xf32NMWf1WmzjuOsUn2cwp/n8Kb6xyAOXYna//AazRDfNuH/oOd0ZnY3LsWPudJiXYehbKk2SJ9t+u/VU1Sjrx3trXUIR/a2AhFL/pT6OTNgkvyvaF8cU6l1MZR0d2XD9W8CyzKofCq1RBX25FG91ksL0AIiunG3HMOq15V/4a5h/5GDf7ukQ8VCGofV+BkEN42sgkhvyRWFR25imdRRGxbWLQbTOyeAZUuGdm2wTtbRJR8MP5E6K8+imdQMCsCRvJ7oRtKDLgnRBLEKSEYsbsF//WRrditQycgP6yXGAF+Tbo5ixhMV++j1BPB5i+h9MV0mugUPnsRDGrvDqfOCeVL962X8f8Vb/GmA0WEHPT23hKD8/5Twi32qeA8h4t9KGRRziZ3jZQVm30gX+vzjGOSLp4r3s8JadhhHDMSAAbCh0G1PKT7gTlPPBwng8lcXWQMwxaYUxSZrsQvQRicP0nk44nBed2ZZ+Gzzrfi+uZmUwM/yTriQY27Th05gMUu1ftLuJot4Z3ZGS1AV+4pApHsjtXjHIWic1HcghOVMag1h6g2TP1MDuBHol4yWbTEf1zG3nn7v/8dfWgqVlmT+ldBERhFR5ky0YQfvRM+0bFPbYCP+dfvigGqBbbkBHondSAeCmdA7q7XskBVbFtRv+FtRlJaQBHQ8T5Xv+SiqFG2PUSlCa9cmsN8r5HAWU7z/lnpOdVY1RcyJ2feFQzW0sLzTli0PcPa1t4Hfzla3QZlo7nRQKyQeXrHIQGuSOdnwLFQSM24IMqsOWun1b0yORp/NFvB2pTY+E1QydXUwSaaspViNghPGvzIFyjDbkXyAt8hrPk6oPFQQNkpCP/W4RvHBlsaJe84ozIOcVPfaylu5h5IvXG26RT6vAXLahX0YudrycUoD8W4NwuwKr5htNrU9gZNYaQaP9UyfGZ58MWf3OO5wPLAVNCHaz5aeXDZ/kwf+qjOe8dCY9BcYWQ7t5dyJz/rnGaD2TcF1ndPoNH7BU1uD9jOTcf4eEvk0I/sGY7Nl/9caONFibe0bq0JE2vop2yHwQqz/XiRszqjjba0TJKQJEvpeIKrx3MEEW00dC4V70EpvEUqHMNvLUNwRnSBbbIj2BkwyyT1ishMLJe46pCAf2v2V+M3HdACGEHB49HucwGJIzS3377DdFCmc/xHo7m24/3npbcukai7aJbqlSdXa8km5NmB2jXYb9XT+0+Ngic7l/3GxZn310/GGzZwffDroPLJa+CmEs1MpJpWXLYd2uqCYLxKrapA/egLYmUjaTR0/MMILi25zLfvdLzbsD8onYc3+GzOEJKpqAjz89AM7e0dwCkXl/C8vtS8DNAbDhJ3B7s1lgweuyesCNhEBoYeJeZP6tKFYouOSiUXyEyRVPP8UukqYbdqGw7kOUJuknMzhKXydmi2tEHWj8hD1DdUCl1UiBqUUPB3Bu0dPj0ZSyqmcSQJ8wEulnRPP91SsyUnDNeV28FXBQzaGbvF7UtSTuoiRtscmt9qbA7E/Istpi5sLDKoki2qYus8gNLNxH7MPpa22ZUp3GaMk7Cus4VWhEe5mh5PElNfieAYxm6FjQAvsmQUWSFYcx2yn0JH2i+FCJyAFYUHcrlYidIXvt/hQsiOdx1KpWTAHrKxh+Ln5KEgGMbJBxzS5sekLQKy5Z/qcxv+ZqppdFHrvTwSDzjdV2UGy6NBBE1/Bc6iad/bPOrYZq3Qj8LfLhMiogDbXnFIfvqNywK7yLzMjGPbq0vLf5lfxWy/gglAn2oK6QoL8eHE6WTgmxrxB0aCwKrXEzzsTCtTWbuMqQNM6xqckL9Kr9TtLVtDyIUwQDFPfs0T9HuQ8hgFrBpfy1cBkhXA1NDQHHfo1UonZRickgT5jFA8JrTdzll+rFMD7LwD7muqqSdGS8V+XcbLm6l1h/dbs3pJGKi/20pHh4GQ9DOr4G4Rd4Mbxb7QxWDXqQ940E8KsXy2YvHKBS6xiRAIbxqKf4Cra1L/kWCayFRoQhJo/6YGpQPqPGYeaooXey8EWlxroxvjoZsGZeo8LHlahnsT3xz8Ka4ODoeQhyS4jtV6eU1oH1uzZeTtGuAu00aH+p0OEpzhBHH+S/Sc+T+aJeiC36h0IayPdDh+xfIXrniPDMqMCUDjzEgonZrRMiXmaNOL8aimM8Rfb9/JmScyfP6BeeQAlG9f4O1NrqGnxh9qsc4NOCxFYSxszCeJDBFUyWa9+7x66AhjS/JjoPaVwLSI7J9v+nfmmCWjUCIG4Avt/OF28e1lhpgllncYFkyX/FLo5/dW+HtmH6eKkhh/bwBblc0/dR3A/cP/l9CN6rE0ifw3XQxM3y6M3U4fd+2DcsPQUs4HOXlvJTvReiiOV2ljnH7SZSa5d5mWolNq/BoJH9Ys2iwBYYPNKkyjpYmch7XxXrT1fdlNJ2O8ZjwkKUiizeyarUDXOJGcUjYRZ2eKzXwL2okZF3wOlmvbPjCu6t4pFaw/DlZZkP6lMARAw/9y79ZVbaV5suqm9JRdQngHMZXUaMePqSYL++oKFZKI9xDvdtY/n/6sdB92jSbIN13r3jrpjaLI1z6Fq4XNkJd5tCWu9gq4ruXzEiSt5L6rrKbJQxAD7YL2SurAWiIrLoY5Mqkd073D1NDzxABKG+c/ZGEhAoaMTpeLncTNer8JkJsgiRl0cGzWk4hfBsm1HHOV2Kl5CewX1JBX4geJYOD5HpTWFeo91tYh+4JBGNlkQYd9ezv8NALA13Mxcf/4KO0FVsxz2aKDM6n+477rXQASP51onLLYAYzxmSaZItNAQv/x10Ox9L85d9K9/IJ0C78/wsI5+Kf2MYnnvo7zc6E4SSEtkc4AWlpNzPCZE0iCKHs2NszY/YxER62UqlWHhmrcJ8v9oqEUOrpaSL5/YAKeU1hpnRHKmmPiNYrWfGK5q9MNDBJGAdP07pVRUoF0Qgw/UdVAg8m9FP9TxMrZ/aTPEwCELzBrVVHjDiHCiQAWW0nsJVhWGTuW+cG3y0IzhILyWuB/PujQug8aHYPvZgl0vXi8fVSkYZBzioHnAKOgSSHaYFP2KqFF/hs50vzTz+/zp1NI/SRzJ79VXslNkaSVan2/wvPb+A5bzjYCn0GedlguhOCDqg4dFqQqL+KNb6zxTiP3Y01cxoiKKZSXysZW41lSa/8pJqzgYEAF919cXKOtTQUE5xeJ0TZWzR43Dh6qCu7M7Pp9V6LMyaMI4tm9lbNOqZAdAe7EWSoTzYalHvnkUrYIN/+5BzCIgK2THy8ZensU0D4r3hcrVBpjMZCV5gkoBk8p27/2aQ5OjT2LVccogB582fc4egenN37MJq0j5ffbJH4I7XTiwnQfAeyKekjskLCDMNDHwA0n5YR17C9fx+GdF13DxfQKdk2Cq8MGeNVR9VS+6VqCKU5h4qDJ1zalioO9HbtcPmf1llOAjlhaiiwHkxRvzR5gSDVRQI+4mwX1YKcfgzu52ASSg0RyrIDc9iZntwpByPh9rJKvt6308IEssI07kN6fD862o+4aXTUJi4AENWjDJExxRyS2RCSd6KzHvOc2cS1jppmzCl4v4o1bWf8GMWN7hXF30NhxrSEfuv+fQp3Uba/XwCPiqlQ2N27JDzOfWYpwxKj8qDUNSOLA1S3z9EmJI7gPPDExnr1bhgVBYPK0HYokPXGTyoGYqetrYy+5lFnn8CNBrdRij1ciwjc74ZfIFa13xZqcR/s5/sx0VlWqMx57+H8tLf6jB5bswmbnW5NDA88KVQTImOINic0JS91H4fMDkmGwxAxfYuEU7oVz3Ol5o57j2RG8cfeHCXRc7AO2ifEaahLpmEtqqYQLP96v/NcnQ82mIn+ryXKylo6J2QzICBJTjSflTUX872VmKhFHe/bdJsc8Hn3dpY6qldjPf9rHUf+rflZ75JJcejpvHifSUeG9IjPJxiMhlY0oVLR3XS2fHs1b32MWtGpA4/hlQFgRJdLeKTtxvdSX1QDG9+ITRkGDws07jNLhsb0YZmOUxbbezy5auQhwY3xQN2Tk1tMGr+wu6lQ4pgKTpXfl3hDffkGyJb/lyFotWPV/grRuCVVqF239w1+JCA+nPbHlPC/ZY2GZL2FFKZKHRRD7e/v7gzODqSlbNERjyoFDODitv2mDloQhJslZ7hpcCZJ7Z7i2GHAJrgU9WatFt6o5n9MnABJCnw8nq2HWaOdb28qnixI6FifDIXzOrqVeMCac1qcuoreoVcLtOqh/PgQdD4RoLSdOPPXkJLkyNxtjbe/VU6E5qYAeqF9sNTLTSyXiUioTM0YkRdM0IUCYNyJjTdvt8IDBBYWiQjr7fSqBUnB0MqelY/qznqK2t7jB3i+t5XtuySIYeizBOp7zXpbs/GSyK6U2HMdmy3I613ngitaP32HvWGBslJLPF4de9DLZ5NVVYS7E1HFog7y20tHrE3p1jFKOQXTjrCrxmLkl0O9zzXnubuKUsRMuutWxZ5j9UwlrAdPCICiBWOiMnEAM9PH2ohyOlB//UuodvdBaDRJvtocx6ajeRgaoy0PR3hLtX8HKL85hGXGsmghDYWJUIHl3QTAsgDLA1a+FdKu7/j5c5eK2rk7GH0oi8a/qpN5W6/nIE//FZc4DuwuFKvPOTGauPf7rwvUo9RT+bx+yMrqB7lvCLm/UecU65pB0tKHztIFIP9ttEKXE8zFJIXtLfw3egt3kgwdct9Gyt/pPgqLoTGnX6SXcR3M1EKhawRKSoZaEEMwBgPkg9yumP/1lS/cqHdcj8tijX+ZfE+RL2xDzs7j348IcaFJbdW3r+leXW0NrLUn+kz/hd48snttBJShmTilDs9+8SqJAejMzwz3n+dxKdlMlRTYsuFQItyxq+wBHjid0Ws8TwPHj5QxJk02QcJ3rHc0dcn3pPkQB6tkpsRupxGc1CRmSSFMRCOzmDb39wx/uGH6y2D7olS+ihYZ52XIp/SeTqNfohryW6F+2WQmn6+OUGMO92bPOoQNRPVCTlUSj0/naWvAHbqc/5jETlgWyHYHZDCWWCkeGHrqWW02Cadz1BMl4UU9N4/Xa38nJE3f5Ycf8FvUG17LUSzRyoIkO7S9SOQPjwCOutXMES5GwM4JpTcmL2DVk7WKXGJxfBN/xxIs+UR5qjD+DgrtL+/vO/FT8lRLld6n00dw6q0deajnAoNh+V+G5GuPLu2Hff2PJHAqaDTSV+zqioETBMiC9AYmZRiPNL71nI/ymwDXS87TZiAjPmQCU+rt0hI9/PnqMDYjiLtHsmwCsk1c1fp1mDWmUKrAKkVZ+In5HfSQdGjymkSfulHa+veaDZB7mlw0VQLPPwL5ga/ClqseonaezSVfcTHWCaiSD9XdGgr1qehdcJ4U/Y7cuB62202pei7H9PF+MfF+rSH+92XzfZBMo+67UTxFRCfHig7zC7zRvBzyK9XIzsgFGIu9SSPFPO1jlBy6EgY3YSMDKMD9btAPww4hybWruhs636SdhjLSRAwSwdTPHbPh2KBVD9V71hZx5OKY2YbyNAwDILc+wMzZqxW8XBK8BHM0TPcYkXRLDb/4PLDVHhSGuZHzrkLqfaWpAv3qeHD6pCKGhQHYGKxr2mYnsIw1qUiMsT4cFUll3Fs48WLdDftGmB2SEgHjBL1b1cIVCIVlOwQtznsEGsEElG336/elXo+xb93J3HP/x7AmH1P6uK8fUXbnckgO7GUuZEZz1seU0BJkFCxvq64/xEeYlpAceNskc1X6Brf5FoJZFkuxyjzIjqINb4AUm310p1IvkqZjuHGhLoeCVyPWBHdVmchwJcqz94nSo6Qr98DCLKjR1lV0GAxCeGGXxaJbbaixXpojBj/PcXTNpkmnsUIQMGQD2DPnQZv9hJkPqf9Y6RbuQs343t36WWYD1w8TLyKQXBqAJdY4LgMsECCqb6/1lS/ej5p5iDCCe28BCjv4NIiwdPDqfBSYhd1bVbwSAAAj8CywQHrXPm6Op8EZBChFNiYzF3i12oxHDzW/OQjHk5mrJihe2DLfT4vRDyQqeiPUOe7XmD+MOgUtQH7X55SS/hrNDCQQ1zVe4OWxIkvS1WIcy67uYiZ/jeiFjFPv4ZgI/MCvRxgbrkjrEUO+XJ5HG7ow0Nb+1rDUReme/e/YFSPBx9+eithff79x8vKW6CpIN7ogGZZjBleI4zfm8ylRLMze1XJgMlKAqKXda+RlcBcHDjjQPCDZrfhipWORDmyCOxBb8ACwsdNgIxz5GzYzb1dg0WXh4fV7QwBRu2KELl/Iiy4CZRFyUTIRlq5CCxdWE1zPqZUWuyztYGqVyCviJlP9Z6EYr2YreKJYM9yqfebcLb+EKRMgMWl17SMvsXhEaw0WeK4hViV26th0WSDFl3mKG/3iAt5K2a40v40NqmVtAlWseLFQ3LM61g864YlNsLoLF/RPw+40R3bB6HRhh6MyIzp3S/ZL0UpHnTTsplCmW0eUU9T9TsvljKo4n5opu1yfzXZ/8OPsyD8MK0/akd5C+nkhpcKYPAKjJ1SZ1hC8rH4+aksTXPZJK/hcZj8QOBQGG3GMGexqPXjrUqpAiZwenYb2uoZACJSIvGjW/4mITWerEXWp7rgBIapZjgUa6a+m1i/uMh9yzmwDn+FavlVxlBmZH1n/8Cp10JXLspjeIobd7SjpNlNpYZIxUpuWmVfYjkLsPmIv64nTtQfeMJmGRAF6wVKKYQlzt6FjuJSe9eypuQELGDeVzXY0pUyuO/+O/TFIfndNFJMoxhanfovSpTTdey3pMK1iFuiIJeJqtQP0GBy3Vf9t40R1cBCmfH1H4BLYKFNfM76Z+ZYz3LBeZQ6YAC/gDHFtRVIKfJoKXtoBAlul1JQZfdSrYMFD72KhBlMfznaeYXedg6aEqbHron2ch23X2olW0T+M0J5cxHFFKn1TyWObdBS9LLVU4b6++5ZYfwkpzv1poFFbpPKRqQ3qI/EX4PxecKFwKrE3q7AGEZjaTaowCL+B+Ce3ZYIMnsl064slysiy4xID28X4pG8bvuwaG2d9iwFrZk4O4LI2pUYshb7uNtOOCme2xOA2Gz0ODsp35IrQo/Z+Z4zwLD7gIogFejdj9tDmayUCsr0/2oxxlaIVVzwZOsubkVsNG8U+OcrgoFTX3Qh5n0WLJ7E/b/3q/s3f5LCws0c2p/+3dSDhjkviRRWr+UcYvFLYffP2EmG9Y6Wax63SyWPYS/At2RsShSjwZ5LEY5GZtByK4Gl/Ak32w186INjgnFCx6Pb+GxUnxnIMeeNFx6pvXeCbVIZ6WA88sQOSCta51TWMUlxbpFfEx+vsx1d6MmeQ6lM7xvAoOybYWCSyTKLZ1dFyzcdLLHED9KPLyNdbvpfu36St6ocXAMCbygS4JwCcPqjoPEW+okOzUjHOt/If49iA/fV+G9bdqLAyLsebJBPusDgi+9DLqsHRmY76FlOHF8/borpPft6bcoQJVkZhhYylbKLr1QCGn+KQ/71W2AJS4pP1qXp0yKbZxkWtxybMG6ehP0qnYuCUZdnZFywIT/ZImRVeWDDFxgA1vBaSwKWkPppKIvX/3sw1qwZKR98YOFj1wmhP4bM8alJGgWUEChIyouImHWh0xOsl2hFlGxRcq/SkechWSU95dBKO1ckAiky0vLdWXPIOTSDBiLQ6k+BRjUSv8APzSg1MbQjT9qKxYx190E0r8A9bYTsTfmfZNazxbopVEfF5An/w3ESySwwa3dklHnLqa8D49CyqSATRjUHJvXcLnnUaYqyFZyWOOgW8CIToZ7i06wCiQs5RRjljF16g6Oa8atJ2kkgsXv5kTOHdV9yIVLrcsb44U+x19V/sYiWAdcrpHYJ5ZhW+gBUpFseIkVy0mBz4ScCnunBnrMTsmZ/agMxNHsE0EcWCMWAGZohQRpZJ9sBDgwAm3n5PpCB2yNc3LMnUZYSJLfliASJsPVQsmfa1i+D5BP11736AaTV+2dXjgRW8IZCHnhFj2DSmLhCRiCsw55tkEMSFhDEDT1otGy0MAAmpXQZttK/QVE6pbTox80F3qtxLI7dO1ATASK6GZUye4kwOLEjWXQ+BXdGiutD7ccX65lL6lVGpXLfcmwbP3BgsG5I2jFcIc8RoObQ9VSOmpq3C2J0+c9erfg16uQ78kk+YO7zaMrrEgGthbMPYd63LH1iz0HO2Ami4BhpNbEodq2ZZtqQply8HNzVcIanmn3y8F0MNY8ES+GHaThCx0kg6LNb5I8f0h+IZJKmBYQa7mztTYMz9SXC8KubFf4fZkoTdNE5NqeP4O3/NmltFc5EwDEFCtWpKfYctucDu0HfmoHKKAkFdli9QTnPt+4ZVxphGv8KNMdBw6hX4ehHX3k3bO/I0OzgEW8h22q5suWBPfdFhtBX8HTn+p+CaRoStvzUMFb+S3fNYzTUZSTUbXqMsCK6TLmmJf0mzIJiisSJIVucE9s+uycVayldA7XN9hFAD+bTNlgcCPPcZtyXNLQ3eqqfYCIDZD9iK8KQJI60aTKtp+3UbEXzxwrHR+U9VzijYfgdc5AFR5EXqnZcK7Z6qMd2T7J+/EqoDZDvsvDTNGOPR6In0BAT30yt/bQgcMNQmbS9cOF3PrOhTPBKP7GJNSotYBO0I7n/VPhfu4jCepaglv/JS5p36VZmsuIz3LJvtAHicPrf2OCBdyaWEKAMUCDaPTksBjj3xKjgkK3owTugAU3ep9d9MVH+mtBe5mzuldu5nV2r+2Mjcf/eozCDJ9/FEZEBKQTS/p2DlP5FYwgdWZ6BNe9tNNVRzdY4ueFw+PzPe+BI6HA8uKZpm7U2xq7a2dmeDqj16BkGfI+6dQd52jB8gUD+l7DFAlRhRMuSrILhRaojCv/OJyW4KCsD3xmGt66DOcmwOW+YnNwkokKm6/qcvHFBwUdkRLcSli4/PcpiUBNKPjZMuNFedbyQQpq5JW11Y+ycEsbdKH3n2WNcvp+00h+vQmdX8N9SizeRoT9RTmWmh4lUsjCzMV3XkgqPDZQSrd2ig4Hupt3Pztg/qzcBw/4sqpHkVvCLeWvQqVq9ypGrI9QM6aVKsAejueoaWN4QpVNA7ndQeDas9QM8UCS2ijvePOWa7dUGTTbQVp36TIT4M7xP+AvHPR1dAVHF+GKXygUJ4C/pHpbWa72gEH/khllbtq7zsf5dr1VeffbWHpvsyfT5bvg4HnkXw/DBNh04psN1n9Igq1cJEshGINBxSXZWsiLLWjkdzA5Ugb8HfFKWZjavLzYSj5dqT12CVz57qcFE43SZtH1UFW/X8SFEqIp+GpQ6I8AoWYgsTy2jvQNBnnjU/30+QsuLqYOW2TMwc5bsEt5sTnVltBZmvUlhCFYcUoYnGvfjShH7zAIkJdsacD5KCZubtW19UraJ4J0AF91YlbIgasCmqpzKsywUhaw+6P1OMX7/FQG4F2tle34dw52bjO0DMpIzg/htxnWVL9KadnOBarXOGDd5aZ83HHOsH9Wc8ZY7464eanlgLS/mpriiWYbeKIH8vt0URgpOm2Jb38qX+rvLwnyV0b+81qqqe2fM6IiStTT0Ce61/Q1P0za3aLnI8Ast6ymPmNV4cPlySvYBnw0MYz73KXlvvB4k/xWPYDFX1KmMpLhFU+2we0LrN391jLDzLWQk1v8uOoDHJuaziMTr4iRAuf6OQRfrwAIlVB4Mw3coLBbBesiK1CbV70qS0osk111Lk/vzo9HlI7yhUw60mmDGhrWs5oGsLGYNJ6EnnDkZhg3weaLRsymHbMkxnUtC2J/IQTJU0I2r7+s74vY8WfZjzQjff+dvPp7lCPh3m0FtMtF+jvApCiimbmlp3Nmfx+/5+csUeNp2A/P8AcrrInoxQhZPbD6hsUzCdEYhnNZAt9FoGlI6QKoI/ayHYjX4ciB+bIplgwrG4KEvwl0cJE6VO/UoMUPK9AELCknyOvHkNAaEOlorC/btYd6NdLVCEy96nZMEdV0Q04REbirN+xOkvAPWowOb4jwPiEaFdrr+rNxaeuJig4o9SMf8l8zw0HQ2pC2aBQDhqukdVOO9xjlZAcNX2OQ37JRYYbOKArdj5k3Y4LOzxByhRD5o5lQqahWwYR/wgLUWpIscpvXYaI+XNU3j6URYAJESqwhl892U8glZ3/+Wd9rfqx36I2dIZl9HNepeG5wntivl3lrmxvI7kZgi2O3g2sW8eomEELaHVhe4QH1U0L5H1b9QlelJSt2CfoDd1hL/KBqI9IqKROyUFKDl9zKwnRbUw0TFX+n0OddxEoCIyFOid3/sAz3XtflI47DMScictj05xZ61MUZ5AXYrqENXe49JMTQnvwFC/SRHHemBtg0O7Wfiuh95XBxBCxKAwcJMYtmYN/Pp10J94x3BVaEblt3dcmL9vPCBH1W8puu0cPXkSmZBjyz8ed5SFYDvzkJGhUlqvaVgVzpZlygsC+v6URdvsJIYxdDEA7XlK4yZsF+VS7a5CzZr1BKUbVsVfVJk31keEQTq5attTadGyYqbj8G0qwAeUVt4xA6D3r5gvsXMS3A9yK9poBYuKx12VAYDI8P+cJn+NynRqgaqjLdpR74L2x764ajBvqK0DVT5UYjX7smqtXTmPx0ujeYnp0YRokHSoNEJKHCpMSeRfasXrUF/oFe7WOvhmQO2tfOzxtkhPRd+tPlnl3bxFhflkkN07Q9ybJmq3KIitrXWkjvi9EUdRLNldSFc39uRVV0HJtd4cAic3p5vTOsrjZaqv/X1UMd5BQ/z2dzcQvrCGY1V5s8iaR0xzVMewy/ltBv0VpnWIsjqe2vo209XQDqylEH3ziSYjDiyt+uce3CEubW6mvsFXQ6CBPkxU30TGSMlnPWhGl/j7vE4SCWeu+vypvwWZ2SwxPz45E7XUgtSgg5CZVxyM5bdSigcMIH0N5+UKXyNTrtQwGxpSftB5whINnJJdCN067PEdWiuiT4+OScpjRvi4ccOe1Eyk3oN7l5t9Po7AtmrQIC70NVlbpO9T6EC3Oz5KvMToS4myl7ZFFqCMjwSW5o66EMmBvOCsBYbuSRIa4G8UvZucRbaaiSjlneiWa1T+ID51LKdb8SSNE5XiPDb8sUKl+apJ8Xlh89D/opFPDfp3ZSRRqtAOhooDd+DlzkBWi/IIUWb63e4lWs0GVzq+DtF9PbfMF3M+rwgahWMSiB+GWmTnYUtpe9XBj7x54WsaViyAPakvZOO1OxYfzJIhiH8DlOZOahUAq9+/Qhwuzv+J4ifzX+dr70EwDGj7MhzquEEmMLLrPt8WS9mc4cGxnhgdemfTiglgClcoZjM5JFUaZ6D+f3hqK2UKUfjAwi9pJg6kfJScgFwxfVb1YzSss8beE5vKEiaic+AVQ1It65gGyjkO2LAE21/xvdRX/cMjKsEEuxazlb5BjD4AMmiZfA2oPHseBrNcOWY7HMIsLiCUvdl1L1MoL81ft5npDg9s0LH4qYnIYvfqfxP4S8rxm/zbDs7nJ/8efLBddW3U8QCoEJI3teVmZpGMC9Od6VIOuLpI15rZemUOwVTNeiC9l60c3BVyEMe8NWTLDIlxiz5W81noZJGy0nVjzl0mys5KLHBJ4kkUotngCN0r+XhdcMEo5JWxvXz5OmILWlL9Nb+JttySkd520bGOUjML+MjU9HXtw2qeTaZijgufM9IyBuTzLnFfjn6wLMGFCLPwzxkFBC06gv72EqKl68bfHjVbZyrFFs1WH5E1IPFuCI+OYquvEvs0Cgrm+ooHdvajtXHbxcSLjwCKk3u6kGEHen2uzL1GRSGIbgmjTIkImNV2jzE6iMxWtDzN2toAXq96WlfK7bBKHHLOSEk3bF1/pkXNhnnDcBlPKohMw9xLXpsF9a0UNdYGr6+NRkroNGHGYTjdRPkY9jzr/5rCCKi4o7pEPTCrT9TnBFk8JkARcT6RkPz9OzrAGX+3va/sUrU2t1JZ6HofSy7fzxB7VDHuZeiJ8cS26JZhKIiR36oho5Y4IxCRZNgzO7u48o8auzKv81xGfGL5ku4OiU9GCQDWAISykJlf4+jbl4o55kz+rxOmflXV43N2VlJbcRabhBZCDw5D2fn2OSSulaa+3UTiAymVAg4COwk+fBXBdj+lfrX6OOoP6MGY9uYUncry2u2cy05epE+nN2KzliNEE6kEF57f/T2Ea+KRJWYdd+mSYZC6/OQJGMytyFGl2Dt5OSgp5i1+dJWAzzpSaMXJxXQhGWf8dx9LJqeiBaMe9YUmfoWvuJdPXZW8FtgGAzqj5/tDO5qZpAgzdqSOWnQPqKX4gUSS9YmYvWv0HOCf+qgvWuwFkDb3IYRSfpkmgHL6ewc05N3SXYRLwKZMU7GBcprAKfpnFSnTKmnngxSXm8u+sP8JQjWb27EOLoqhBLH/oeP95iFcoArPHtLXwAwRCBgUqDgDvVaDFj8iKypYmDdTY/YBub3VYKK5Yb0/GXz5V7ER2doTm+dvL7Hy06DCyleUrtDZDwZ7Myumup+3yMwsdb8iCyePEJeHoO+xeAJN4BYZ+WfdKGqal5tH2rgBRPwuhrC9cWtjyjOphirh3eh5auWOzLVaqIn1gkJ3aVoJIVjoaR6BwZzhjxdrIu+WbeBG9Qzkobzlsq0lHTsrIeB+zBCDGfBx2LbqbaD9s3dS2pYz0Ua6rght1SmlkU+AaspvRDnF978JnFg/xNpdCfJiP4OaLP7K1szcHWieBPcgmjyKRJ3XaZTj3Kq0JM3750GWLTMHMTun5fWqmaeKCUPKz/SZyY+0W7XGrc+8ruiSjAyAxL7WjE4GDqd36eDvW9OFOwUM349Vghbyys8msPVs3Wjr9+1BhURp280c7fjWp8balju/QHLPDltFe2Hm3pMDSQ/5pAPfo6LHWZE7QLidaqXYp/UERi92ximsXY2ike/7bHsn/t7gQBMgsLXnZnzd9osjpKW6MVX/5MJpK1OFPnJGjxn3dxCPG+F38rnAIWsrVDht4hlojEVyc9mBenO7/45VE7jJYH6zRpAwluYLcqeIP04Wi6DeOHdYTgxMfPH6NxQNIR1WzTHFI1C03q/8P3h9xD9eWpJZhPM09G2uzeZf9JhlGJ/oCixAK15IkPDBurlSd0TwOnJ4cKqIjzXYRfSYuLwbsWMUNIoGqF+C5hsqh9tQ4TsP3XoaSThVa7BATZ7cF2v7Dg7GQ4aufKaJd/xoVFRtRBYoHpqojhqWll4wJS4dbYwHrMtGXdh0vdkm1Rf1//wX2FA6sGcy5PW1pbbWV7+Zc1HM2dh1zIQJKPDrFaZ9O/1BaqXVS7Kp7r1TWMKa2c80RJfHGxz2E4Xqx6kgueZFjLaUIXw2WIawtf2INMEoFxohFTbYxX78Hmt2HOCLZc+d7+QXa1pEl5IwYhR3bzVSH817h2sqc2EvupBBZl9Bg62bwdzlrtj1hQ8+cD9X3iSm7XsoVw7XqsWELSLtEFAs9OxEFRPqN9lL+K2zjjb0MuZelN7L01O0toyOHCM4lEbzhmLZ7nG/ftLXfhDYMIs/Edw0QqF4DdKQ6d5C8/d9dOewcFg1R8JQRn9/yz/CZe+jPdLEZKc3mztoLRI06gtdG+PQjwtqNxpC78pcy0z2qKPoq2f/IP4Ui1mrMzA5ttHVZsdF1GT1u8SoLPG/SBmdUsSqIp6/KCcM3czW85PllQYJxskI7klJHW+ozz+0Idq8KxGixxX018ePES7DNXtn9iboybkCVFVvcrRDzM17Cl4wvKXMOZl1qcAxxRmwds29xjcum11IHpUcgaYtb1BhU+G06jrWj8XU+Q69zsjo0K+7Z/6raqNYbfmAtsD1lWOU1vNLr+GD0A2sD2mIWBqlJqTX7zxt7Guj0ooFTv24/zVWs6LDb3+RbrKaKGUV/TbuPlPh6tg2zjwRKgZeGV0tsux42D5g4u3iCk8Mvk3n3EtsKvlfHpp+pqff+i4MBfn+pNWza46+piwyyQPh9C06i9OzmAaJjtuJnm9BlYTJ28syjmGrs9OAn7Q7m0JOsdp6WVEr+CEkzECU20y6bHHVwiucKuF9SRmofUSiAifipCQa3+4ElTe65ol+TRasSzyIli7eowZo1C+A2i6OzmxpG+1SfzMkubdcbBAh/JPJKPrvzDePdgA4O5PJ9n/9hSOzCtQN9x1Ysxg4oAz+JLLICOrS4ku9+OeKviTY54ovsoqLCJONHMRe0a6VCYhGwNScxpNeKYjLDXZpRD3DZJTr1pILfJW48HAihSiC/HvzGnY40kK9OlSoV4GD9nvlppGE6PeBfC5iPPypJ1WyVf5fu24RmEW7fo63tFfMjz7/UIErDVerEreMXLnakCOj+W12yU3y9mr73rL0g5KeVxww1FCSmX62f4/aymnWxn/8f4zd37cGnZnwd2cBkwb4sq6mVT4jE13X+ZYAAmmhuWYHWFWyqmbIhzqmglnCJokkxjfYLm+HyrdG3bpq2vrw3iMhpIbRtAmIG9wlMLiycpJhLMcFZe8yjcqDV5RIkVTRaXU5fCwZ89PgCiboD3LkOBWS8Wvyv6QVxYn4JIR8P/4vcTdQ550+7QuTJnH8RXxN0q0YgkIYVhes3bRjeVfY79Y/fYcWd7M2H8DsFhBT0qNwfZ170qVJ/dS4fx7jmIr7/I6F6xmpe28gOSRq+WTG3tTO00mIXWKSBexJqR5io1rQphIvNVnW69KF+aDTp3M3QmFu/3EbpLDZVLma2EMvEpAGn3z1CU5Hs69tMH9A89DgUspqS4N42rpBa34WmpVUXkZVkDbjTjsvum9FZNOQ70lvUhDdqfbnjUt4qr/vDhPbS4xz8FzJYq0CXdV1Cr3AgEYtirTNhJyA+GSMv6oIa2gWv2NV4UHjpUr6jTSQA7giOs+WRIpYiKIWLyg5dFYYftYPIDhwrcHn70yDdeMLfICjXkCAvNzpos95vZnPClc37ACyTEo3fh7P6oukVktM9z7pUeSoGB/9V6OL+Lgv26LX7hfpgFfOddJUDHq4EoL1bD+58dRrRwKhkr54nqJemFwpyZQAs7GkvX4QCNQgWu7wOmxAyHHtjWBb+swZBzcj1ZOyPKTDrTS9u+bOKglCbIhpb6Ig9GxAu5bX+67qLkABxbFlq6necZPlf6/K6Uj403C3GXnASPw8NEOeX6wUaNZ4jv52iYEqtqajzpgufNC45xHKoMHYiD/hEDkWLtbxuEJ/rxzPF5ssHU3gLlB96s9AnK0o6yKExl8caKOeQ0pZzEgMSVEIL7DIpE2M15h6FtFtCvQRfnZmvX1pAmSrPvsJTnHbWoL8O+XYgvabL2sgkkMyDSIYyUW5OVmsEYsYtulYe+exzbdrXEj/vAkYb+Hxx3pRHPVbzjW0HlEMQzsZouagxpuTtr+aGS3ZOv/3l+O73fRX/7QP1qS4mREYS/Ug+dNzxV/aWm5QLihqknu+djSNkqOFSZfvEUU4/05IyqaeDSX7ghEMTJ9Md1MHbAYskvl/aDtX2l1BKn9RSg2M8x3mBuUegf8OXVHBxPnwF64fnh4stPyGfm5SagorPlMY5eA0ykgx5p7rKKuMqogAzX8XT2YvXJ9Uoddf9mSm6dHBe9a3b4ea/BFyaG9YG27eRtf1SG6L8XIOH4Hvb0aTz1vFXn3MTJWTyFsE/sMga+EjFFX1sanzlEjqHed+j9zPDqKhuV8gfDbJ1WoQkmg7s0gvav6S4qSWwOBEhGfRnb/JXsDAgSQPJ4M4cifpDdCnLA0+QAIflhyXK7sj+y/fbS5iBjJwIfyWfTFu5WMvONBxcqIg+kyujMOfX1lFGQbU7Ogh3ESekLnwIxM7Iyrd2jaCk2//KOgbgYnEMPDGPEn3x8ZX5qgnI+UkEnfq6Jf745zbDaMRxH8xqbCF4jizscGQgkmrMcpk60Ok8v/IwhnhuuE4H98RPNJacT8CAU/BhZURgZR7XT4lPGnO1siwLXbfECU10nIVidcC/caFvwLCU5aCAyuWhJwWj3AJwWpTMeoATu0eVAT21lC4S+5SYLEKkVazi9mZPfzxeD4EA/M4g2x+pqZiZVd2nX0A9Oq9DVMmDmDFGZutZrKq397BzYiNDH+y/3lzmFbwVlaxa49hzGURhypWQ7ma76nxWpekH68iZVeOkgwGpMuX16+/VNcHwNSDEElGFzNwXiiIokfjfTR9LvqjUQ2NwjeavGrK6n0266jUc/1Tn9uPxlvaiuHhJXLNDWAo4ajUYRU1S6IEQRs0VEFXD/YoiOjKlZRZcQ0Ghni+SoL3jxDtdAyX0RjJBhBA5Cb0Dwx2FyEjWgLCiqrlS6bqAt8/KDeyR2ah75men+EvRWnhzWzH3atYBVqhqge6c0m5pw8O5BZMeBIHLvn37lCIf9gdoFK9hb1Cq1qdkvbEpT8sKQg2kiBJuTOJ1Cwy6chO7KNwX0In1yhovYvUQq4z3rht4a7pdVnOqssNvgbQwnm0dXDQHy1fCun4avYOzTeIgdca7hGbm3CA+ENNDNeAjetXvTyOvXt6cO/61op+0oCBLkOAp2d2Zw8zsz5DeArL+1l5oi2yBvAmL0ZYtKMzyBqUrj2eNV+M6QEN36f1QrBcihiXOOprjWxJNta0sZ9o1m6AWzYPnVCynAxfrN+ysImy8p0BS8xFDC7t1NowVPRMygRei/yFxB0Yi3/Sa6DqSjhJV4zbAie2+CRP8eWSYJllOffdRvBlXz30wCBMp4Ep0Ga+5dH0XjJ37sJfIFOrQoJuT0fPPrtUSZoPJ9unZkSuKSrxjNiWKi0dqGlQISlZj9CAKU/KAJBO3ReqCnuEjUR9Nyng08UedL9HYggCzaXkyaOstJ52opSmO7PY+bHa3KLGLX4G2lf/63vZFELO3BFxmpZw9R51rOMBF0tQVUNUToPZQuPflrVcqdP3/hxDJ/GHxKf/LhKhieDlEuvyBhaCATgEyifBfg6DHkg0HWuqrXjiAReoRhK2jGJbCutFMz0cV7CsOytiyVXvTSIm6yDf8qXnqAWYWXms+prSlpezJ+PGjuAnQYYlE1HI1rfvIWR/QWFG+oxSmAnGDdTBQB7EbR5vMVA5LxfUULio1oPjs9K3iuPXnDvvnWyYEPjWL2bgHfZ5TlIM1oVYBvWzjPVlobrB3jqJ5MyxvobDVNVuqX679yLFW1HEt2kDFp77PStosD0sVChQq++i4TFScUwogfUufxRrFzgyAWaE0tmpJ9uj0vSZUg0JCaSlbAxEDJL/ahcqQiyCd2gUmiv4t2h52fQdAzCXCnDBxkPlE8mBWDPtXbdWCPhTtPHJdOYcvQ7LUIdRqfijKX/vBVA4SNzw0NDcLTLWmGG9K+2ElOK9yCVCucZKV8f+Cp1GfeJyeshXbs/TCYzCInSzyMB5eqFSjoaJS1S7uLo8ZdANxRE/mbdftw+iwdQiaGWAqkqAbW3awKNIUVsTys1cezZ1LnJwgvWb363XNvvqGQhvr9sVLvwu0E13zxx18+21GdwuhE7drK7dpNiZoyRQynKX1Y0lOqb3G0W5uFmhcBpf8D2+HvJTWE58gHUVR3RCntmFtwe2TEdh7iHgjRl+gu93IGF+aIwW4h9YEh0yNCNxb/XocRcXHhQyrCYAP7UKn1DGf6qHOuM35qbePDSkNnDNBZxIuoY7z2aKYD2nLaJ+2dZPFvqBCSUpJZKz69cCy1vSzfVL/ZMO2XqiOHaoYzAd4jrVaHQjHP1XuieOE6AlMn5RRM2+01KJ+s70Pi2kEiPVY8QTAZV1oKjJKYYOeOTeywTHUe4+0MqhyW7SKeMR7H5k7RhpgR4jJUq2QH6aaUf7tPZHsjm2H6ISbHmstXN4askqGPtdXw3ixCDtXC7DOo/r2wZ2RdlSO96v1AENNbEIeonjBimas/7PwM0b1us8yGfLJI2W16+jf/+56paN2e670/C4W1pihxu7g2MRUyxme3MG4066JzjWh6YGa9s7QUaVRolH1/dVf/j/cgyN7hijOIDILEQCrLsmXa6aqY7ttej1IZcjB34q9uE2o3ZgciCDSdQ1NHKmCXsUb6V3GvTBEV3Vk1WTqPEls8WjHgxoLwqAtA83X6RVjGgK4KOnWa7D19sCedYyjlveeDP/OecqPTYNXEzhc/oHhEfr6rHdRD7UQ3bfqvA2F5K8eBYM/fUK4k6a2MnFG1Uc2oMbpNZEM7OJx+HqhqarGIr3F91yBqWQbU57yyjNNv7hfOg/A0OSAT1OeLInNS+BHfzBEFVg0w3KfpvXsVAMA3czQU8z0w5r34tC+EtNg7lk72LEsQhXMjq+I4HH0E3LIUlXGJIm4D2r1tfOQYVws079s9KBFmKoDKeHm/cD8n2g/3A/DXJKoclmB6F7JnrmC3lcCOzAhqq6lw0PVzOW8k3t0nWM1LT3YVPmRd0MBhwAQWfMp51tM6qcrtpBXQpF8iR8uPcSPK9ILf4B8tCzD8WfA7KH3PuGQicIl8TgSQC6uE95RdnDCD1iEm3ZgZuU2zzdGuBPFuP3Z9B578sE9X5KSWrXa/8R7zIzVpoC30PHsiudb2AScdo8Sqk1nPgtx7mhVlCw6vz5LBAsi3Llckspr/9licdkCZkTGmYAxZny6lJ2z9DnUk7qflzSib4e+2rBZMHBEEm+9gwC98v5q6C57Vfzdz/GDXPVLVyjpOQmTF0943iJnOh2uvMXqM55cNBNomMY0QkkNvd2UNZK2Zea333/TbCakmL9RNe7V9teUs2G2N53cxAIwbjJ2MNtKLWb9i9dyic/n5+/UwC3TSkllFa+D+6zI9CwOHTmlVBWy53Cp+O2xk3oj0whc6s4oSBbgo8Nfhsfpbx9pQBc53HI9SQD4ohRDs20tovk8S1t6v3VpXoa9Zx5Gt9ovoXLPCCRFdMFCliZHKKscsnhEBVOv5x5pcybyLmpkJUix5+KuN1NywhwV2eJcq1z47olJtYPFqOOL/kjdydVmVUrEx2mkVsHTEcd5ttE1ilk9ZzIewCz2hRxa5bKdKMVBMFimTjBzwVYsK4Pc2jCZcRqWfGACbMP+xwa9hHpLv7+u94RfJLhMFhxfiBUX8Y7pe9wLIRTeEOi94HcgP3f9CoSSHLRGTzTUvbBJ3H496xSVSeCa0RSCQ52PqM0Bhpf3c1D0fMpG2SFY0OEBcpP8xBfb9VnHe4RiT4c5zQgkv8+Xxu9eYce89LtJUBWWcLt/UL1LbTrEEy6NZ1bCzQv9PLp1OQJFt0/xsWYSdJ53pasPr27LKOwCmx8lsOxEAUr1qTzX8IkDiDXCcK01ie8GRVc24Y6S1CvxIa+mTfDaQwc9c6VecwhqFyiHzNrT/+UKY8qsUjPox+KBekAeW/6nBO1MOwvlUAZrKY+/S8gGpd3hjXzr85OamluSp2IA6KMZHzYuiz1Bts742kNa7i2JoUio3BH+35c/SNm3QfmgawyLVqqvb9VmYnXkBKOijWzZm0BNZ5Lnu7foYE8m+AWwj3YF95jly93jooyiMxYpe69rNdRTIDkaPabUkv2QItZQxNUOFY0YeDyXmbwZxpoFVvL2CI4ri00eKaDeIkesJ7cZUVVlf3SNhLCPl/x75sGFUZlcji69ZNgj1mO9wTuZLJdrdIg8m0vQE2w/wiqQxM7AryYbBhhtfnY3sAg0Pb+ciOtFpizoeEcp6xYRzJjCVdDq5ek/TR/uy1w9K+eHrrawZkuP7pf/Qif7/ykuG42HjasJZYDv6wIxDGjUj/tljzUC/IeQ1pL3AG78Fs5kBbjMHFpOJdfP3C1Vt8b0KUkldBSQ2UDxv9DsrF8mlNYL7syArMsbT1SIA8c82Hua3dIa/VANvvKoGcVeW4X035rvmVNtAOh9PGb/gJgndkhqUmk1K6zJruzaUMKdY0rNj1TH1nk3vNoMbkwuu5kuk72b6819LU+x7pL0Lvu+6JR52nAQvaLgnrOmxnVkvT4qcadwiFk+UWzr5xaSFD3qWHl42Ykq+Bnyy1iYUQOK5WG5ef6UAwlHa5IpIW2lANu4JpzaI4OkoONAKZ3T2wzPC1lqAXMvlCftCI9soiqorHE3Tt0JQ3uVDbfkYUsN6brStFR1CXUOq8duzkMF8KRw/lCtkOhpcN20YTWJccF7DGIuKL67wTtrwh8M/pM/7Bo6J8qRNYv3kXWOuNllxAHXtZpBFjYDkF/vAUAo+FXDsoS3jBcHDRgH2NEIIk6josNKvFvAXRqeiMAyS4rP/I/GLKD+LeV7bN8GaFNMEKjjTrGmldhTRDKgobpoxuJuvcD6h+7IYg4rogq83y68DjvNUgmg5VP00euCjqPx1vh4U/4AJ0scBcNdEzxeZmK35kca9kx7Hpk0Ic+r2XTvZHRmYfBEC+vKDVoTfHusHx+jdJSwqJGNycUMvcw7mOHBIkWIXcnTVhS57E3Ep9LaBPhmS53Z5A+Wq/uWIofGlKd0eqAFJiACooNxZ7coyRuADHOZFAYGYKVgmsZd7RWN3IqNDHj5aWY0bFBmIFY6byqY7UWxNufZJPpCrt9Ji3hbkydhRpt/9WSTEsgWAs2R+ZV9HLNQZ03yX87U8f2hnW5yCB23fGnFuYeBlw4zEersXu7RLw5Iz/rjN4R9cz6xJCzrTdZRCEdFKqMsoIaD+CIb/5W69UKj8IUPRG7OXuRy1+fTWiyNxbXoVMvPc3xRUoh9Tqq3xSIikIYrtPtngB/KvJFOrIaI9VUtmujJor+CujzfdLx3WniD+e+Iod2z9Ex0qcD7iQE9qb8TMDhUVvv/rEN8rvGJrfnxDibBGIna/SzCuUUF/uEbxk15zWjXR7xJ8VHHdFxnOX4K+LJGWsfc14mrtzst8C+C3MxNJi5upZ4W0OYo03/CGp6gLUlOLKhU9q9TTyAfUAHyl6XqE/JTU2aj5mnLVITDVjwKj91pbO0EeDyxsfIJDq7va1tIr81mswMAvf4NjkPRLNwvn8hvvdDkt9ruELXhmH9npfUS9lLksL1/kjl/gaTJBjmi6/wYP0ai2KdZaMbNEyog7rex+VdKPNR/at6ADMhMTxHNywK2pJbIydNHH4tVsRubnIk6x+xY3bXUu7MlWz4cQHpWeghzMjGTrKYd23NXI1tN3L65HAOwyoiVhd/xm8r5eFFmRfR4WHF6mLq+r5LpPCgLwLPnEq/OkLhbhPA7ski8vbm1yanKtEPOEijIYdhDaesCTQ81L7DP+OQfxVIBH2VUxLzileu8uCONFYzb3sZ9TAXmMx2k58wrKGBe2glZw8B/m5mc5VNhKw2Vo3s0omrKduJ4f6Kx7qS3l2RG1oXJ7mdzWQ9r3I0jSZkKarESSqb9D9g3u9QffNhT2wQdpySdbTxveNSdcS6Lxb9Qm6v/GNK7fD71ZGwON5To622I5iD/dQ8WgBEMl8y0lSf1NNVYlVJmO5mytD4w8O5+AzHBzd9V4xriYDPdylh5XJm6KWXUmFZPNgojjG5+eCnlVzCaHmB1bA/Sv851ywdCa7NkWdRPRvCop9fJPP+LjBFamDwpXKtY7u5Jn4fa8Bhqd6kLZLlziXB3+NqNh7g5DOC3r5wVFdOk8f244ep9PYSnhUnCGX3bzKF7q1MHEw8daVL+188EqRSPRhF3AZE0TnESXGtwxJprRLfJNZkhk6dxt52bzUO+gpPl86m63HmBE7Ki+AapXOAmJ6gO1/lsCfJ+llu2r+jgJh/V/hU1zDTDem4cs7xVNnLs7toOCDp+15h/1iaJFpuldY0HJiiovcGSoZi6HhroGqTTQbCS6rFzRpA3Ci6LuyEDeLjdqK88Ak/paWgfwyvujrvcFf9Oxu7YOrFqzlG6wTnREVRbD9R9w/vRazB8x6KPRxUE8j+uU7NMcUPr43WIHpW2dNJrBZpchgPqXoxUmbRqfTvLFZsrRJVGvMlGEmj7l7wPzFWXO6pC/bQ0ggK6I0OP4pdFXx+wKza0bykEVw1sfjgxALL1lD5gJxyzqrErF4J0pfaFDA8sxoQe6oxfNleGF4oyew9scApYTXFl1/d7H+ZQszopjDSEIJ6J/hQbchkyExiYzd7uAN6OWECZuDU35jl4fBuOhxm44f4/cyxVusrE82+X5eHXg6P5xLui7nP4FmhySsuH3hvxi7H/B1vGAphYF9I9VHhCbOOWiFtj05pfLPuD1dPZQEvH/fwRzsLx+m2LL22QgRtbb6fvbEeeDmnqEi3gO+89Qze1PV1zbVwFe1HrqCgDvOIMjSF/TCkwcj2ajnygBbwn7HecORqv97r6wmdvFxpjWb5nzThbUf23WmAwE1XKTOWAgtgpbfU2bxmxLn4VsTd9T2LPwOtjZpS9TNO7wCKowvo8Tw217OpUvms9tJ0MBB49bc+ZCpOnic7xqc2z4SqyaMh24eQUKeQAWH1Oz2973ORXMvMVxuLZwlnCnijVTFFtA/VrvT3KirVQhiC18nHFkEW/Iyyy0Smn+F9chECYTUVqwOSzxCR5L2r1rLV8y+JJnKX7CDsW2nYvhpx4xkQk4aVIEDMGRbeOJRHjSzVxAGk/I0zLUjvqWiMkigE6an8jvRgsJIsjEgDRo5vbPnwL7nxQZMj8OWDpwGrxtma4pU6mOkWrVMwEZi93QStto8l0FsLst3Owld7U/Zc+1ZMRTVyuu9iN6YaO8cAdJ8SPqsJggs3gzbh0pSv1jEMW4TOGd7GSvgm9vZKoh7PecbBmKnmfXqv6r2tHlpBrL/XjLMZLpL+LcKQ7vyYqC0IqR7Al9+nmMzC/EA8xauq3VlsSzKtT2SLJ02h+0+wtMkqDSoCoCugEEPBxvfgYcT/ZxcyARAlelo0UzjjPshv9eBEO6ugz8MZI47r6O+SLVIlh62T+OyPyEZEu0NUk9B26zZ24q7am6Bla/5/r/D+PFL/74hPXaiMF+k7/+FQRhQ1YtIbbRfqv/N6esd8gNnmH8KGkbAl4Z1nfldlup1QjMDV7eJOgSE5ibEXZL17qD43KUcWO4AW3snAd1xYDiuPRHQHChpJHRMBKwAAHhGIxOfKcKcbOiYt3qkhgIZnY/64KUinoSz1iFWEJpta3DyRqq8KrLHYz2Q3KzWjoauQFldJqiQN7wIfc2skFMFV2Z7m4CdpZwVjxF0gv/W2HR+A6ddvgiIz6jN5mJzMxczEQFZLH4gk5vjZ4JZWBHeLCWriTPy1rvevHpX4LDViMuOkFrD8nLe1J4ZtbtS0B0WrINWAn8R9Ge1Q/vcMGNupmAlUq42YETlbaaXDsoM1Yg9qvxiLSpOFkhZtSOcNd6xeYxmxKGE167CXS88O9uwkSqc0lxgLE8H1PARB8G5DXJA+dGU+QW+OBc404hmKqggGul9N/jMGRsRNV8hu/cberaS1bdFKiOWne8N6KOjgIx5VFaRZQsd/nR+3Pc45N3Fkw/qlC6Vbm00aE4nkE2Dh8Twak31LBdgEKth9xIrR4bl9T5/kkurIooebJ0+e6i2JS01TBt5zmJRUxfLen4jDAAz1a7WUORAx4+SReZDMkUMwMJrdPkHTvCe2vylVu8oUDEEjskRNCcUdPNbk34adPyOydztRFwyKDPEM37IucoCdZw4PlEl9WFS7MbffORPQn+eJqrjyuSs5I5WMflxqBoscOOpFvz627gQPdbvhUc6gqHIfWNqY5MBuxBTGVAhbqwFZaOYpXzw5NBWuJS9ac4SK++hjbHiCc2sM8E+RpyDXzvnvUZEteNMhbdb+jrnfsF+GSWo+PFrFMv5cY/+AxZ4C4VhK6LAF2lMjbsYUuLiRcasrmwdeuGIHkvQHGwJ2b/+Do31B5eQeEL1i1wj+NQhLL9IUleNO7zxXVcw+mTnJkCDgmhL38Wo+urQge52yTmZ17tz5b5ot3Oz7+PYtd4R9/D2w3SoCDu6nXwV/tvJVl4qIInNV59PmmCzP73MxkpKiEtSBHyJaNEPv+A5FcZ6GJ8uDtSQJNkhorTxbFvPYVWyNdpPJIFr8Lq8JqH9/VoLPeHzln4HzNLLKqftdi1ovxsyF0tSM0dgMdsJ/oK5dZ2kSZXkyI2B29MSfVdKWIokao36EVhBV7WKFXuVU16HP8iWuJWKqwV5n/3MhlIPDDyGfpDT8FkvuX9iHmd7O1FIO/y8kBuPeAYNUS7tDoC/air7IFE7ZJprTDl192UtoINZOtlO3eaCodyBn0Wxrwg9md5ksB5wqkwTyYR+4e9egOoY/XtTtrAwI7raARdu2HnDg7NsbWhe5ZyHjEt6TanwZxOLfkSzI/PmOAlUIhLEbgqovUyj/zdQkaxA/Z42AfR23QTI2kQEZdqAP8n6LhFpCR3JZbPSaAVq6Qa9K0wLMfGQkEVxGmKS1TJKVTQw46ZW04Y7YqS/TpjiL3jWXxi48+CpAqC8mV+wyDRQuwh0o70b+RqUzo2NGC0Y2rqLf6nl2MENuqclWqIMjZsrlMFbBm+zir+MJCKk1qXKzyM4Wdg7jylDLHnM1b+R8SiRbrnPD7QXYcfHUwD4FvQMdeEUMfXUHfTPuPhqUNHzst20mDzmIE8z62CDcrhT2LkLeOP/esoDH8q5UqMeGHIDaX+UOTqVtK+22yNrOMjb8XKO7WfuTyYmhJs0EZh6oszN24pbzq01qTeBhv2nOi2UAdy49HrEWcmqBRaJN16hXeCLm8ptBMThiOf2OMyA7N9fAVf9H0iK9LoMPf6LpPGoknuHpLle9IQc1TRaeH9G6nm0/bbxCcUmlt6fNXaPJDnzWpoepm5hNSsF/TNJ7VBNDkAaZKIe0tIKa7oKrYaXJcKd2jhAurh8ynclj53iDjosPJ40K8ZtCfed2xNrPph/6BDAeWp/yiz3NT+yJ+tlQ8LxH0OTuPR1gc8c48j0arRXG0bcQV1yLwsd/VLH+Rscm/CYtIfidSGsV9FdQlYqDkru9+bKgX4HtEWMWR8TwDjQCJFqs13dKhkfY0MMkgfY3LDST8ozWEeZtnAjVC/byaMeb2jcMzoW1pGYs0C0JEeRntXH7BVVj2KQDfGJnN1lP7o1G74viGzOAY7sIDK15w80YXQEKPnA9kqSCa+z0THUlqOREgjRIcbcmmvk1x8eOeMCqU+MQkfj6CpT+O1edSo01EQl6ALzaHXot2i5onXZOarqr+dkHXUr5NQlK5qLvQmr6BHtBvN1M475mgUncpmIkAnh6SfuQwbglgEt6/CeI65YZCZppZrXyn9qSEhmhf48X46xCJWpvbUIhPIBXSAKLggKCxFRQCYbNkoGFn0f1rTMeHrY65sy9eqsbWYEnhExuPrdqgZuuQ/RpKXYRLXKN+uingygDkfGL8vRKRZBawvVBYcZZwf7YkNWFhnGeleIXx9MPC4e9eF9xApkmSkQIcM0GhR5I5HOymnGy7Z07PIFFPNsUWsHxHQPVKafMB/AsS8KN4UiRz3pIQR+J1l9jRphJfC1KgMrzST2b0QzZHeCbz7kElB5mcP8MyRUWxzW8tuANYe+SSrGwrwV3v0jLuLov/xQSrQ1J2fGXcASM1fsDwAdWl+/YMbAqwHa2lU8ZLkDRngWiZQeI0aO9YtkfgswyMQxaJUBF6fnSCXfEKlNvrw9u0aySz6gwbFLuxgTjVqEBLl3bGYhRquNlvcpcN+36cKqdxkEvVIXv3hqFAnO5m/F6Zvyx94WVNaaEB+H6dHhiD6cDMYl+lXBZ0B6f29ALQyPIzCKXLhWtMR8wCtkipW6HzaL2vXOP98nt3Hvb9qbeivHMPeGcpBaiw8ow6GqpSaokFsz7UG8oU0qFYc13vm/QmR7vQtnZrLvyWCYB/tXGgLAn1Z8ERcupwOQulhyusZdVun1T9Jfdlysv42C8ucBh1J7RUEjSRWOkPFaGdsKW7YJoswxtt1cEtf7pT6ofX2q6f1o8nIcAJzO1aazsvqs/piZpjW+l7w2O1OHtegitB7xtbf5y/y2sHdM1dikDfDyL3vJqJ0HCNjEvbI/rrNSZ/saIKKh7P2w99N5AnnrX1lDDMwYi3Juli8aG1xrUJ5uTMF9ZVzFb5SPvlRd3xLHFljqOnCjVIcfVFgYTgEpeDELdnYM1g6qj/RyiOuap+VGAQfI6uz+a4IJcwKAY3/waq1HbbfUz8c2F7rRkj9rEgJVcJO8apJW32bbD+Blprw7xUwYLgngd4UkCSuCo572kWGeQh1ieOzLq7nqZm8OGkk6mCYMhbwdav2I60Ts92s/OSzua2tIPrTwzRvbJsTaRnHINStZcj8YmpdpcsexcjALZEGxPck2MsqP5xtuSbQ2uWxX2Q75fPVoS+VNdUMgP978XVRHrgrD1hlmYhV/Egc1D+juZuvntlEJcrK3+/cBn75tnsbOk2da5huj8nVYcaebC9Aci8nRkDwESOfiZnDTM3TQnmD4f6eNFTcUNp2pl/ez7OWbWAqWCXjT3cYfy9k8ZbktBSPEinmFI1LOqR3L0axGQXRcSgHPPo/TWfwyz1fxCNUuMagGFA1Gd8LmnXK/GJwCMjAm6jXyIe20a8riaHgYTcOgSku6yuInD7rnRdn7VkSgxplV4HKbn8EMwnVhTtMMsA7ftldA76gnM58DZngsYRtiWspz2IVjwC9Vh+vmctFBbuX4OWONSA1h/7ka4h0ZRJbmnFb9597Ye+AYc9rCu5pAkdWutgxa5FavFkKk4rDFbKasve0/lCMhtnYLoBn7GE/J4kBL9Gk8va2jeseWcjRMY8I9K+H/Y0bckUPUDXl1fd8OyOuxnJHnqoF8JO6uhkHNoi8+H6huaRFtvN68T7Nc1GPBwf5O1cjhFbBGlpc5Ipkt6Ar31qu9f6k3DF2X5S1E2Zy7j3KsaNWuPky/1ajeZy6CloxmeiZaSF3jXGsnFkoFMqwH7shugGCuXOEwzhNtF7Jzi5zMgIGVRWC/bMl8zP/QwIQpLjNGvbrzptiXrvY9O1W3c5s6lLTi+TdPjZ24rG1k53C2Ak7B2+CzyWTkABg34GDAxOUU+Bpb9oexl6YIfRnwehZ0QL10v1/6hsyttZwTR7HT1G+tOWbtPnXMoferQTZa2BsHzKN/BcZbjI71d9G0AC1sQnnrn/PBY4P0TDQx07eBEgCxZGuTBxQk3ioI/jrNS/1r2yeazmqDUFDzku1n7FU/H1A3LIlsDLYZU6mQunK5SR2fFeTSHh/wCpCy4XCDlVffHTLKQBeBy0YzMu7lKaZpWtGYw6cjc4fQ2iMjQP9Ai66WotCV8vl58Q8mnUPHiqIdlBR5cd5JOP59E3Vt3JAarLaPgwYtqzAEnEyQ6z54u2BbPDyB4vaOtHUs+FVrzpyG2VbseZGNNLlsIrUNcscl9UH40h6ya2TqUiG2/16O2bS1e6KcLVrvUE393mSk1vK1uLPY3g+VBBd+jTIya5SzX9GoqIC/cLB8UZ3AuiA2sVo1a0QxFA4J9b05G9qbtBX5Pdjcj/O4b0n/F0qCJr7lIRfc0sXrQpMfPAcp7AEE8LgmMor+vwiLPQPg4R8qIkkuo8j0Fxz3ZI5gZOQAUwCSv8RbJdCYKCHshotlNOpqr2YJgv8cBjdlJZFcBi3rg0HBq/lw/FS45uLcSu7DHQyXmVL0WTvnSXaqTyheJqWZPwDucZDimW0esxhrYgNP0ww2M6Ix59g4vlZVUrDUlWW3Qnp7PnrZi9agoYmAJMR9zSHEWLgSXUnA6dxIcB7LCAV67ypIGChKIScJwgtjLyYVyNVHptLWZoWDSnB773cwYEZUt412wA3Ttburr3o89GDycGzWIPdLoVNwAJ3Y+y3MsIsYJ+gfSJZsyIh4v6X4hKd9dyEBPtU5ECVhDu24T5mpyaz1GCaj6d5/tpywvHe+w+v1TTihInMSuYYJoUh3AL2maMnLcu/omBNjvcnTU00Zcv/7lki1QF7htNoNmAs4kqLF5OHlu13INOpuMbRnv7vV8gO4xp0kKaXsueo22laK3IMEWXFNJboACZrkzxfCpdTpq20BOXQpUN5Hl5ThmnRUq4X8ZyzSpe8eURrToHzHBjLf3zUMZc85RdtbV9UKb/RuXJyuw/BUJmYKaVV85KGNraCQjomZ2CxH1ow89m5CFuUmk+yLO1JF5mkX7a2C2FfgsjuScmyRTmRFBAwEGu2VouK/W3Vp4eK9Vaylwjmim7J2fPtATF7cV0CCNqmRFA4SJjShkspgAeVp9NacQdmITP3QAvb0VxFufoG3q0uRQ3CFrR3PESbKLXQbqig6S79Hv4JcHp6InqiNVIPM6/DFh9v9uerOpyRfhDMeix9T/+wFBKcBVjgmIPz7K7Dk5wCz0YmGs0nGjJle0qeDaZ/41keTFOuOMVDtXEweZb+j+Fx8LHnyWw9/pOYnbDqDmkn/W8Iar2B+7/+Ot6PbIm75ra5eTOkOWW/7FcJJAJ0BjtgkMiU/nxLoc8nUX53uwZVMmhUXbubDuEgsqajnqMVX3NiSMqh211VZzQQkwz9io0CrC77prZRaj1PHTAy6RWkTqJcfko3BLuQwvfKdS/L6mOe4CVrfBNZ7CvlGcn7HsTWHUUa2FCn1bAQlSgD8Mh8kFNVj+8MdQQMJ1g8dfQjguhyRS6VYdZQp1QOOgFNZBeNmCQmvPbHvSyo++JZMbSnMm6lh3dric4W6RvWEfki0lBBFlCW3Mkzu/mNf19hX3ezR8cZJgwIXlkJLSilcE5sgeMq1Ku1VhrEFRVs7WXKPqNYFFOvrTXfHverjYa5vcKmZOeIpkGaLL3gWaAPrkZNGHcyVI0w4hd4hgCSaC2jbXkrW4dhtWIuAwWSNPpYuhK+NNpwV7g0KJiYd+3mL7Viz1YOiaMLMjvL0HM5ammkrMQTrPgmyJYtMJamKWGOiByPSXkyepusZ3DE0WcFy5bEB6a0ZSB0/lyZexCXiXTjQhmFv8ZMXY5nEQay6YbFxPL2tI4z5MPmPkg5bkZeAjDZ7X77hErKwEP15Ls6jHGRbwHH5b5MqPabRgtlxSHhvBWiu9K6JxthSM93w18LCFYKgUnyxiXsI92dzhF5NKqTJllKprudSd5KXVR7oV4SXsJiG4yWsBuD/Y4u6xRZ6t6+x9L3/Ce+pue3XwbfA8azx2OyaLktwXTQKqkCRBZzmGRJ1YfbNuN83FicL8L9ZqmyqR+d96xoIv1Pth4NJ3+vsOs4DmnEPhTkZfH3WkHlHCPviyaOY6lvWgLkr+J4RqTGVcyBtmOGGflW9Ostbs5B+BD7Gzlu/axr2RvUkrqsllSdyF9NdYwY39pBAZ1EHZlaEr2HapcWFEW4StCMXK/MvXKBdZhRuzDGGUB4BBMRpp7WDO37GpqZrPxZ1dymISVT1pMYRJ/z9+oayR9XwyekeT+P9Tv2eBLsQJq/YFvbhskoMIKzEgbePL6LitOX63vSe2ap0lhVJFg+k416Pasucl9zqZ5oQ/yNbTmFMLvJWn2qEc/0/xWfAbVzOQoHcNj+MdwbEqHyXsVdkumJhk5wHvwmYohw2C6t4dx6JQ5sNVzXf14FwE1ioqkHFrjLaIBeM1u0OeT1kSH0fI/CzFx2SRIu5HnXmkBrskKClgmgSpJwl8p+1F8RIjyYU7aAlyqim9Tp30izlk+ARmvtjbJT8JLZt9uRgaIDLrNj/EHmlkx8EQee1GrJ/v2vruPbJI7/3kIy+U8hbSj0wsEnksxcgfP2dh2MEVSxuLtRNB7fEaVtAazrXgUosMAueJ6GEm6U8PT6tpdFnpuVuCfKxtEEsDnNUDjmwsmsW8ai45qjl8fw8Xa/y/92falyb881z1K+EA0evTmuWdgh5pFZlX9CrMo+cP+tl15DmLYihw7wTcSoY6SDJlDiHCrjuUfXWyFHn8LZqui8mg10WrpE/aLytqJusS4DTqY8BVMu2KR9HOYNtm5YeUnOAQXhuHm89rXQ5NHXCfKmBxct5JV7jECTC3nIC9TnSS4PH/MZkIhY48SaFytnTAZqOxxY8lUClrzcrAiWbwSBgfkzj0a6GXIdO1I4hJ2TJ9hOU7SjDlNs+IjBiQBwcJG3Fm/tzMYjQa2+3UeNNbK5g5zBwZWWlmL8rNV5Yz+m9y1RSg/9lwg/sQRJo21zSrSr38EtND/T95IEnsfOpGk/XitmXUtiSNGMVmHj7V9bUFrrkXqLwD+hX7dozEeb/8xEug5n2xE6q+arkTuD3gxilXE0MbkHPQnme6lm4Jr4guh0b2ytghepJAcWxX0Kdi6Du+o5uBbDOV+hhQ0COIXYdgIQrvtO2eS7S5oE4UBG7+q9aPhAFytEI2U2V+ZjU1hS0pjFuXUYo8BQduYySmeWmgT6L2Xnw05rGMKlQLaED/Bsm2bWiY0qZh9VCRuDDQUTZ+irkjW9wDqzICdyd/NqrAAeTa1umFHnj5hf99h6kqVdBEXgA/3i3LQC2riwbX3tWEIgntHH/DWYlLQdctwriIQeBRwWJXhB6QWPiWbegoiL6o0r1udU4HNyLFPkTdKqfksXoFLxYavFCsezxEaPfMS53TO6NiBHgq1izcR6bfoP7NDkwajQjJe2/Ob7Ow34FHVqs7VzqgNU9ZqnLYibHTUpMVxGqFeo4ekK5U0kLO6WpkApr4ie2i9Y/Qn5rIeEAQgmO7z8kAOj0vz3yTHGV+2+YCTqYl7J6If3+3j0/Kb9vsZbQlIQyzn+l3xzXDA5KADYps4z0a10bsWW6U0GBDRlXSJ5VhBMmFUffXvI0HlIcKk6ElhUcWL/GTjg+0K10qwvrLCzmIEupgZpVez4ojDKNXVG6cgzA9l0g5HDZY9ykuOj2mkuO4CSPpZfx10Cb4CNQyICQg7oD0+hzKIiNLqDIWDyN8oodddOhQoy3CseB9+Y3mU9sjdnUykxtYWqMBsauOTJRrV/t24uOmC5/kXmxNcf5IG1XfFMx2J3f5Icj+wvag22XS7cScMaHB429xEyvemwkNcf5gQESs/Xx0+AohP8Ffqk/R6dPzv2exWvNYSlYIi7+x2/VHPu2gYCNJDF2JTiWPZ4PWBFn4c8u4MvLJ3bBxOiAF7mELpgv6SXjdMlfWg4THJ7GYHt/KT6Y5oXmXFYjoFV4JCE6csQ1uo6YMQQj4i12oICOOaQyVxrfSTjCDacsb/nwOwqNsy1Tz40QY5Tp40wjdt4grcnxmSl/7iJAwAhi0jFXhnJ5M34uBSqQ1G3dQ/sZxwviUYn/OF0Ol59eSYIC5I4gACHLGS0n3I6jwkQQRLBMrrNmz3Kel+4hCvkKYF+ZwpNv91Vbe4HXbh40BpwW1XqfRoPOyKQUQjL8G6B2KzzsTWRYH49DXT2KeEM4rdTRxM9zO25xy7NZwDW5WInM4alyG5dyN2WgbA0wC0XtKJ6s41TUJe75bSRYIAzx2eH1B4oEdufg/Kmex8pnJnrxt2tCh3Koplis6Btoly5k1zkT/HQ4q/8PimzHJ2Z/Cdptr5r3GKZ5cvYoJcffI5O0tvMemBZH0aGXfAGWhc1qipdMagPDV1QtJGuMzYcQaXpCaws99FQKuWKpDMfYmvY/pN/ueyenlW4dqsfXEHNKxpsMCNsezaX3WMiXqvQuQRyDCmWgQRZ5QpnPIN3DSyA6wtj57+cZi0Onhc/kvl4HcfETdK5+dFgKD21zuKJecAuX+WsyiFBNhCMp7MYaoQbgVGOdcZ7nZNJkYUY21qUt9dqBsXJbAg7sw0yTVc2wOdaM3Um+1M5v+lSwQlcWez+OXAv1cjAMCV38uMoGjPSL3SmbsHaq1d46zgEiOqfl+5kxsWYVzm1EdVBT9gtPaoVdNCf0o5PgmT2CshCjt/rEe9n3Ih7rOZNeIHRIdcAoCGSYiBK6quF7kaEHAYP9OTw+EL/w7cCho3/Rde2gu9iFhh/dj8dHUTSO/d3veLJofyIgXzbhd4YKrrreQF+/WgfJ4mugI1RIPl2p8r3Kw69pQmEt0sm7p7ZPUhvza3FBojlFq6aUYcep59eY/ZzyKnjdWwm841if2LSziQ8Hwj9JF6vp1+Xvmb1GVd4ehdxrZypuSLWb3DOIbw9UDR4R9KmrpZjXrKDCOZBWKWCsgCeTBJo+S350V/oEK21gn2EWVavcMqIP+o+ssZ7CS22K59gargahil5VFSFPf5gsOAoFSBltGRpPoo504zUM3Z34xz/BdWpaPzkVxP8gkFGusgXyt7K3y0CiL6tEXqFq0yc31TqRm35UuzrzmsG7ZVM1xjQio2xAxe8Suu5qNKNSj6IlzazhHPH4o4yMNabuAgpHGzsT0HrRfEMrrkxq2naZPC8g6M+5DH4wEIeYmxWeDYOicwTQa5LHUyS2cARh3x3wgLBTFL14rmO1zYcWOk5UhwdMaiZj0Gmc/emoj/dAjNSa56qYnYPQCkfQlFqy37H00rmb6ld+51n/pRU+VYXZktsbcNyMQ3oX8sOeHi4tTOr51W+/fL4IWbQIDwTxk3n3tEcEncaKZlLCbYM1nIzT4zsb9PMbVws2D4q6MKO2X7S+/vZgC5hw+ctmcWH1tGDlOOoRgHgJnZFBumfWvAQCJdhvpa16p+7mTq6E9HPOhor2Hty8hpFbCVmdDkyGRG+kjifxWmtdbIsOerTpkkVlMNkYXTczeNRzhfOlIsGqNTrPff0fjOhxSxEVB9Z/Zikgmm3YbmVk1Ot7VlALMvWM44mJeoT2eZOqa1xalXPxALKufN+8g2AmPgG5+HwSWLEdAlYhEGoh0bCUpswUjqzeu0uX5a3S1NTY+MfEjVKA5nvDRIjxm6kAO6yBFZ5mjt0r9NkWH6DSQFCyKL3l4+uSXqVMrHb8/DJH5XJLsNaMe4T/27Me0S3rT7jrjfGqU3zAPBuItMVYp9oElrOAgcrWfJH5nvt5bAGwrGYDUr4+8VUUaySdM4k03iYX/oTdByXFFkBrfNomTxYOyU6daFp3FPkvPXF3tA0UwUb63ewMa1PnxOr5uNLHWSZoLdVrOTJyZn1K7NyK3XU9XgWU8MC84YlJTgOVO77dXY60cWhdXdGfSrxuaVZRfsMScGlMf6AEE9sqYDVH6U2GO3q/ALUZkF64Q60lPgKnHQjM0GSK9Y9vGTGEkFwKNJMprC2oAf9oFqbr/nXnWK8TR+LJY+d+iHE+bFv3g6OLtKoJTeTIZwW6cIEW7pZ656kQtmh60SRBM+qYD8cCLAPozp8XaR2LQz6fWKZVJ6ncC3nC0/9q7UfHz1doYCE2rdOzrEVU/6qPpgHDYVia2Q6YHhv8Si8RW6lqV6XoWBN0J7i0QZ3hk8+F83D6bx298Zvi8YGd+Vno0CFIe3y7ARvZMKHtGcOYMqiJCmPSnnetlEPJ/+1cwGEHjPMvMASyBC9u419h/a2E+0a2ZySXf6jnzLxhGjt6pp3h2zlY0az0DEOZeMMiXZZCCmUkLAF7pQp7MqMlMuMnOsrkdbClk1slFW+vnGeicG/okt9vkvrsRCR7wAAEPY8olA0AB2Wjj2ZN2AKmEmeTemFThT7WBdcQti0WxGFgE0gbLalfOvO0vOOsPElIjDsxvWQkYN1udHsv0BrAHBdvDxcXgzjhYWM5jiQ1B/DOowC8uq+7zDSgCRrvBYmZzKuSCmTMuCB1nwaMrgSdLlXfJ74rIAczUPuzyS8dMDePseeWmgUprucwjToq2NTsDJ3QGvuSjumwv7ubxF8EnQHfe3+GyGgv4qxZIJsMaUDchlJKFIPrRU+9rh5DDbgu1BVNNR40urhG5RcOG7JYbx7NhWVhkkFMCLdCWeVYS2cbg+ohnsM25qsjy36I0jkn97VukHcuwNH0xOoK/E4R+SGAfeRe4tRNw0+kiIGY9YPTJc1fvqFvqNSEYeBBZ2G+BFTF1pRpFyLCrXQ56kPbzW0kcWL4FwzX9XZG5jGeKc0HK8IT3/ICt6iENZbIFHYVK1c47aEZsVVBLtBOHJ3N3jSX/lAN8URyN5pVEIKKcf2m/hOr//eN0ooDiV02ddLSCduk0F+x1gBXxB9u5IsY8ZlEVcoIa25LJhTcvmlBwOrh2nN2QnD9FCN6hBZzmOoj8/EA46Q2Jrb8gOWOIZu2twhMgYfYvcc41kw2kZzr/u0Tqi3v0tTgUCh/t7frevxaCvViXQ27u2E3BolE8NcZi3agMf7dx/KtTl2KedubYU0RoX0NWkWxXeEjTdW36GIuxe+hkTgE6GZ+yWWnegD0/laFPipj6zXcVIa0UA5jqF8ltentnI3b1rvuKwFNz9wGTxlJRuwnu8KMpXrg/McrKKXFVu7MYsGjsRsDY1yuYmRw/5EonrlZgYuCm5ljs+Q/THnv6R6aw2PBI5IeaYHekmfYHgbrzmXjZNARLBm8OeycwSWXIu9NCUb7LHpAQtZFns4eo5nWAdGE6dZqmBJxCdLGkdbSVpMYoe11G9nTkYT6U3l50TMax4mQ/n7YJW+P5qTBIhkE7aR2vWDvapzfzrmsIvvALY+jyIy8Y0PCoHZYezk/CIx7KniqKN0tZJiuvwdmjS6VzBTvXcVKUNQeplp3dkbDYrtwmRG1WWTNLH8drK6LNQpQeVQ9qWfl6LIcYCYNzoYl9460LDNUdSoAChDD5wxxFvi0mTPFlK0CN4si7YMOEqts9pExv+akv+OhqNxPZQub80lk5fDbYfSJzCVcguks7c5J8hUVl0M5+9KolhUd1yNqv1JPmZPTzhX4unYnQfrzrnGqgBIrGAEFwEW4bF39mTuJq35C9/B6IAzW4QYw1Rl037ZJ0yODlmm8YXzZ4zEBRS3V5+qW/MvOKY6JQYYYPxnnvLsD5lFTwO/+tdG3ek58X+PLyD3W6S3dWf0U0os41x2A4HGEi6B0J+cKrocqvUyLz1ZBubpGn7OkFYzRTdCcnX9vPdBkPnk868J5Egkv4vfDhonae5vc3A1I5MozeX42fMWlDsuUhD1QViv/+1kzBwd+11AGz+i+lpOADp2vO59Rc/H36qHRpS9RDhDCSMJt7O/vPQfuNmVaiL8vc5D/abwldb5HUfn+ihN535uIfUpRXDf6qF0TbR/SeUqKWCFnX8bWvCuyrF8npEj6VYGR5KpbTAwer4/FIxzMoO3Gnq58psH7mV1hlsNq1deKZeoaggeUf6/MiW66/KIsp2+t99pNNS3cvM0PhqTolGc4bDfq4KwfCpnIPwlpMYDy8+2Lu4Yy46DvnSsB2gxiEm10azgEx6Jy/lyCnLgW7awvMg5XsjxCpZxhKdR0Rag9oMi+7r1G96zG21zTDEtz9NdXCDdHKZvIt+1IEOKTBvdYJ94NBdWw9BY/bA/XwyYxO2185AHL47wQJt+TY/HTzWfgQll7IaoTs5R+yqclg/kcM/Iy1wp3jvDanXlahJ8Ph9HfHbfWotfU8cVH7poF5fCLDD/9xpv4Oq+F28Y3otxRJGL8cKS3f0cz6exseulAwnfFFz6TBZ8NLbgU3+gfHjdpQy+Hl09vfdwP465o9nJyaSGRA7TK/Q3V7+rQwgtd0vrw8TNkPEYFP1PyJddmIHEIlYfOmK6geLpElw1pdMyNdKWPwcNyWMVyybeldmBrfcS7YaBbVPPXn5t9cEqi/CPT32GWU8l49jpBc37N1zlmyJF6yK/RIaWZyIRIpD4Hdc7J+3vfMn3J5/YIh4D+SajoAWGGGOb5OxOYn+fwoRTKhu4CMn3HDo3BeQOUp0iSCKdljeob02IjSrzstudSPdj29Vy/juRrdlY4iWTmLuhUYStta//8D6jdGiHhtXbzjgp/yyg0Zaf1oGkI5jApEMJTpe5WnhjHJ8ZrBPzq5As0ff5zszKJz7xb6dZU1kohxYBDCNZFp4Fn6TdFIhs2/37HoNOEiNNOxGFKT3lTucRii4qClc54xFYXhcKGeo4O2eu37f81Z2uf/GoYjnofwRns9mzwo3z1q66/0UwOo6hgLfmmg6WoLGd3XT8R3UiwaYlPOeFtQK9VUge1fudNIs74ltA85LgKh/BUrdy/Gh+0rbqVa7XBWWwxFabqMRpG4H37i6t1kdEozHHhs6Dpisd0lismI20EN6nz4TrXFBs0uHGhAhz5xRxhShNasVZMV6I01/SVKBjevwepNzj8pHFX+oE6jeJT1byVKE7//0Hfanu4ympOXPhePm1wvvAxSYOy6EF0ptWwU7AlfoA+aQFXAQwNFezR/fNh6KOhV5Q01zF9hhmqeIR3VuqMsIO0rTe7NhdaDKSb7YlwpyITxstBUgLDxqY3rbCUEOlsvNO5nGxu3mSRNUzznzJv1dG+zNzNyiK804Hv6Fz63yrfSrk6S0KuLhwVD94KqqgWrOmvo116zgzxeYnddE7KgB8It0y5tb80H9qWW74jW98XnZDAfuFzsvLQbNONSIs4NCp4ywWsYlJrNjHjikAi6ssiqSmqDwLmbtDATHtWNAdf8zDNmavXyIW6vTBf39I5k7xijZ25TJvUjqPIf/BGjTeDLcEOlGon6DrjyutFD2r1OtJFo3N0DRM5yzjPN6CyNfdnFN6Q4hhLI/CZKtyknMcgl0rfAiwMXwLV+GDnHZtCjc9t213u4um+ZERJnNGEg2u4cH4mCi3bU4O4DW3TuZGq+CSWUWOHiQoGcBV9ItseSKyJ9IOmfNHlgPCGDWLLgcQGSrKt8/VG9c41FNkZVwpiHDTLUJorS/3rk5rn03u92hNfViTyezxlvNDmQ698QyXQs9Wi0Dvw3h3r8XfKO+Q9Ua+NTLCYHrAnUtcPrhxQv/LSQQW/EqGYWCp4FWanXpAsfpjcQMxpNighmBagEzqeyDb8xyFOBsBh1wFBRx0vDsJh57iGIuaB7/Hi5X/YpURQwOBAQpvXVdNtttUZNYi9OCpsIt6FQoSXs5bJ4+IdqmDDHOHvqMz64u6kkgXHeoiwtM0O+SvNXUYuzbe05iZD2EHvS0V+Wr94r5Ni60aUN5hFZpFDGEo8nIvrL6cp8Mxbxxb3qgVni23UtV3bDxveoxkRNK1IcLJHgE9sHudTEg892SWqJgl/wKgz9YcI9xrigQWn2Iqervr2vvIU+d5ZHdBaGgGW5UY1qDj/INemYFxqx2D7i6jOsTYh6yuWstzBy0e5eteiKLgOeaPVDXRe2+foLjfbqZ08I70EbAreRCubZhe7Gusu3wQ0dJOtKrnRTtlfUyjlNIbUw93dKZNqckPQNMrFXI35V79k4u/D0Qc4F5HCNu0xkFSTIx8nuSSfAzxxBMpHHjQXJeHEZfF2/RHrzFWlmrRFI95TMh+xcDBZ/JWDQNno4LFGOZPv2IJZe/UkeTF9IyTMybySNV4v+2GEXxcrlwnbFD8+9cqvpma68W02xGwSOAqkR4OZghEI6EOSwMZ5xiRbhUn0OCuJH/WUkogKwC4v+HNXjp7swLTQF79nCR3Scd9ZTHZ3ko8/Gf4d+NpsLejvx8nIubyQRuZnX6xfxsDVg6o+Mz76tGzo1y3BSDRKXYryiGmDOeqo4foeFd6CbXdVYyiJy368k4TDlhWVacbXMGLEZjTh7RJsZNVv46tGsa4JtbV85JNDTxv6di6XKWC8akUvbEqm7dksZrdzgCIePf7ac1TH6t/BFb1okK3p2RcYKth5v/PkAlOkmUGW3YBsx41sycdhfVsbJ/EnV0ruYPQAZKggaQ260R+9IBCHaJJPJHW3a1eDeGCP2n+uNNJuEuErX/uFp+lfgRMmrJ7I3HQa+drzaVpqz4pUklJmbHNGMj/ssLzPEnwoWBt+1AeZ4DIAs/IGcBFl/rpvxuNcbuRTNoDYXOXIgD/f0tuvdBHsLRnM6uCi2nh/DZ1KgdxRELigvzEuc8Yze8ZaDUG/8ZcReKJYiSKidGK9YjKasGWDux7TGbWnJgYVylnEIPjuYMHUeXwjXFeXUHYtYvrNyNueix7bNaW3vbiqgTj0HKkmrVfBMqVZkb342Ke4LlF4Avd7Hu9je1lj+6vP+YPu4Osm6K0DYUiXPDJvGvH6hVJLsETs5PtVCKuLK2CUTZJ2Cz/pnz3ABr/tzbb1Kxh/rr3aE+gsjhd4jo3T7IufgtP8RQxbO7vledinlTupEA+DrW1Z85qwlnN8f3A2B0qeh7I4jRXFFu7GHvGHJtXemGxts1R3lmlxZ+zZYVBPeGNvof0hJ7uMePo0hbfgH/52zrKJAofimIagDeWnLMjLxuoZQ2sm1HF+kyAHAYnilAcfjw4fYG27RYwyZ0sOhJdXZHmDguUm1EQiaILhZISw4uNTy4Wq28rhEtN41zOCooC7Mt8StHBA4AFENVGGmpYAo6KgUxFo3s87rY1ad5he094xwcGeWag8JQThtlHGiyb7iDist8LRjyzsGJ4x8NVTLSmPXSNLjmEUmKjrBnznBswOcNs5Bl9N0MfgqyGGAHSZk6oQbb4C5VDiORJbjaO8xrAxKqs5L7bMauhdQKaVN4NBqd/37Hs8pf+iE31LOml1pPISA1imq9Q1XM3l++fznnKYvRe44IoPF0iSHBqRO2yuRUf+0izd4pW/YBHiq7fnWfiLR+TdWhoXkKkMg60ofYKWDmUGm0gewKgXV2xBh3ci2YQVo5xAfZCUekqg4wk1So9ISYNiftUbrWBppmOgCwEpLJGwh0xVZGvpTrOd8zvJsYr8hDvTF2FFCkZLR/Vep/PxFQVb0iMHlZvf8EYzmzIPdqeeDhfTnOcQCPcvDRcpsxpZonl47coqzXTDBIsl8AekPbyELnywnOpj7p1vDVc2GqLuhElfnoBFGNIcvpa/ZF8JEGZJli66bRexruKa4wS3GRRFNLZwQgTEGNXJ7B1zijEdL7yKBhAnFpqEI9F/rAquKnjC5BqH+oSeVUevrHbnJCP09vsVeITHhzOrB379coAEFmpysPXyGvW87yMOsQCGjxQhvrISJXfG67iLdFlWxzW6S7j5xSJWs9x45qom+zBlvIRyFxDQdvofa+IG4mhx+XRQiKZK4FAJGVy/sCGk931xUZJNDdj3lYHxNgUw9DSboZITRHJZ4HM5c6OGx7Har4SP29rsK3hBsVDWbtCILanHAP7kAJEeOLFOcEOUzxXgBoynCV5xiEU4K2sCAUe/1sGZdSgv32mDLxYclbO89OxGWGrmnzDUdA6qYSncmFuHCu2Si2GGTkGtrzoG333t7Gatfp+IOyseg9JTMbqyuIAM9MfgF8+lGIWkPPINKP28JOvdJia6vZIKk+ML6P4QhH0luXDuQZ6twOQSebbnp4/JHEl+IXi6XXkIAEQgN8IQBX7tM9JBTGm6RDwOpZkQYNByjnsj4PE/47M20DyiaR+4jdNigVPH38SIO7jl6dayXYwG1hjI7bqSChukMC4L6n2qP+s7kMQtuRuThpbVYhFj4bm+UoIcD+8XfjkSAu9wVhHkKD1vxAGqQSPHdNwFfr4W4dlBRNsSAR4yKHCam0HL5MzJfRYNSMYlCTuJkcVhfiZO9a2q0peUe/yktUq3RezCVpGv70ovUyEqXZBB4+3r+95h4tCjaLjmUahLGlQIBcgEqu25tq8ZmAhhLqo2qubiHQX6/oDKcVHRu0yX6687hBIuL+KzYOVAXzD+Flhws4OnNmtimWrbpM1NHc8eEqOc7NJ2n29Ngkl935w44knMmvJcKuvLelJUI4IM+fW9TSm2dKTZmbGHzdd8Qv/ZEjWHkkHMMhwIlYXCpXoti09x0rOjeWb6sFh62e3ZSU85aFX8VPmzGwvZDDogxrAeq88W1zhj/sMuK1+UtuAmloMXjs9c+8UfjtakakIUcSy1PBGxxDsEBqXXQoFCIPZkDaOz4FxVyHOFODUqAvAyod57DlnvR8AdFDB1bgtNqG85OxX5+wxXrFnFAVVQeK41e4pFn/LpnahqcvanF0zbikXIa4pSyskEPTcHR0lQEwfqLk0a9+73JYmKmGq/tRjhr6+kGl3F02chLQ+IWfePF9neNyv1HwCx72rFO8GoYdx//0rUGyc8Si50YcVg8XfHifSIhtx7ndEFcoMpiF0Y0QfAliY0mjVIjIiAseJsdzTmUf7wMufng31qmGlNe/x2CNdu0ldFes66KwwUtRaULKs1rp+BIYW+vUa0JvTYhDi/p6/hk08becWkS4dEQuhQFOgfAWxa1UJ+5xsH6MBuuGWqivxpOMmirUnycxaz0bfKHtGPhM7ub4DGyn0uk0aNaoX6meAEra+zFSzCqoHHd96lATk0yLRYl+xYjqwdkEpLeWz8yGJbCg5p3U1dDR0xYhkeBcvFpb8w+QRy1+3uFW91j1N7NNIAgsRn7srJMckvautxI6HdSOllJ3gjTZcxzDtYf7JlS7sLpcFnwrJMhLVbe00ideuSqKf1M3BKnz/QgHsju9ibHsWN11xRHFcq79n63bq3EOHiGQXv4JqkWzvKIbkldQ0e2M4+7oGyPfXDQ3WsNDCTKww0GWeNgoSAQtMDwGasmt7DwpnTYmSZhhT1/ycWU9o21LZlSg4CS21MhM47oc2CN6tTv6i3xaCzZcGH9anF8q9hUmymkfwyX4yW/dxKGiHMrmdJim3pJEZTf1lcBk4Q6zAiKo4mfi2FtZOM0mJwXZoVhBWFvUfcAfm5wE/gsVUAFdQ4Ztl2pMDey4FRMKYqr2dRiKzpfaeyZEnvhn+cZgxbUsL+CRQAvU1PkXKtG007GhPI70+LBVODgaf3pmezdIWKmMJ5kCqTSVEbj2DComltJooJXwSCeGGZ4YpD/4D3HuEuwbz/up2NrCXbTu4OYKSoySUq24nO0UodyZ1LOtsQGvOJOIoxkFnR264X/e1ylUnSK8dqbkIqra07hUeC8KQUVedOY42cs+mjHbAFWudQLY7DkS17tt98AjcNDLRJ/nJsDGV+MlxvGW77f+i2JBNGbxHZNR9mwkyc0NEh+bpDttkVY5pU1hgqAz3IcR+phn/JCah7/uT0JTST0ymp/bTGgrbDbwzkPpll8M3mAf5E+xsUGU1gjLZUGszKBSLgfWjIJHNTFA1KuDL9t0nP3D5T2NOiBtd7CJcQ3Vvi42GiBXjxceathCuCMk09M3E/v46aGYEG0H/xXoxT34eAVqduvq2l2YjlSOP8JWmIpj2xG0nAIn1IAM2+T3bfmNbg226y+yKR3PDpCykDGJDZZXboBDmhEoLhXo2tLmvdtG7bEKxO1sfS36U4lhp8sb/rQtc85gqIL+0vyninDm2UlAoSVJ6/kyNmRBeTluqXa8DDZR0z2Hwtu1I8Zijixv4aYKK1ROj2U6cZvitn1QNvTFkg1ChL7CMfcQrDNXEuNtMrM5NaQy6LYqYFPM0oaIjkD4FLiRlgRvwrs64au0Rie3QMYJQLcijVLHW/YlxZ8WiHv1cqoOcO6ccuYEqz/bIbJo5h9EeeE1LfyhgDxIEEAh7yZGD6nqOOzLPPtbq9rJzzfiJhHy57feUe3FPAgJkbPsWAE4LhPuuXc3JNBjzSTU1EGdZFs8Yk+T/iFZlu3jKGpUvLC/N2BFyw7Z6Z2ZTaMQCKXRvGuP20jU7EKovPByKYk4sYOO6l0rgiXywghhlj7CGmT8EwL2z0mBe+CQm3FULWizv2obQLtmrl9/VmQ3bcXRUwydpRPhRxSaQ1uo1lx2QyyI0sX4IerqelT3RecFFbBN6dM4BemDAynwKckb5RGsnQ45bKw0Ytj3sBnfbs6Jk7J6z9hBpHcS2+D7wQtrMr3mDgMN9lXnQL2r6yhYnNcYG0Z++80SGWOWNfd7CoYK9zRoFAP1ls5SrJGMSbVUEc6pMv5I9XFUKIe0J6l958YTOfTkK86phErg30NMRjIXschtZfK/VOlEGNLbydUSKd/QDHpqCjMEModufucOYDyqj1nAdyJZ8GHYN/xqzgct0OQg5SmYyMQlVRMkcHKgvU3jPyVXeZGvtWtPsK4CYQE0gfwSqwumkItNjeaM/Wrs8C0dLOzUWWXLMTckGeW2PN4y1tP76nRby5yVHm/vMuYStkkvzxPaDvi1bU5BUoA73AhlTybwUDV+d7cyTIe/d40oxSiJJrxmIsqBPfYRBEat2MZnkFHf3x/UJ6MKpYPQzH2Yu9M7g2elmzkAoyPW7DrD9YT9nPdH2bpFFc7mSa8V3li/Vinf1Aaur2TKlWNnmRuJ4tXlsvIZoyed+ArpUiwbKjsYvsCdDV5Wl0wC3W4Bxdl8POadQos1HUwkmXRrGCbcG+5ZV7JYGxOT1szb0M7vd60VeSh5Q4+J1YJL0ytQLcNEPE0RbGQO48G6IKYGe5HsHbqC3gZFR1ztwFKJSDiJFSE1iQUWNFFAyvbZWltWOyYx8mWZmSdr0CRmnh5kNgnfCyo2noLHSVM20Gd2E860TgKnWioIYIPz1MiNdtf6Pj7RUv/QHoIUR28jUF1aO4HwvbqOs4fFbxUvhDMNu7/UPypmg5y5v91Pxtk1U96xBgtaV1N1EenNb6do9ekeqnPtCpP4EDAj9XOvDVR6RRK9FiuCeB1KkNrHnimDDIsslYpbiXaVIIH9S/Axqj3DtnPbZ5PuY8SkEUlyEMV2ZxmI6dzkcUywvUVwxeoRk29zXCm0H6B+EB6D8hy0VetsbNq18/IJJRCaq1mjhpqd4f/BdwfPuJ8rs0x+3/vr8wKv9CW0QkGQJ9zEdr7Xq0EaNW8UT8KvmP/ig2Yegl6q17ECqFSEVVUTG9soNAoArjuGGS8yHWPppvQ41UylaXo3kV3nMpJ+xg+DKyl1saKv/ucQ5XSn+P99P89Wcc0GLELIHS1WNaUB3SNO11Nbax+AV57hKfKYuZTEHPo1hg11SP6wjzN82Tzn2AkjHXqz66KT+znC//hBRf4RsTVm/j04YVWB0BwvkTjV2Ce06ypWqiz6GyJXj1EahP+rp+hXQ48uqBvFoz5JMtvrHxTpQuZV0QjWX91GA1BJqgKlNDjOqcePk2sbSoI6lz9pUHB1keXVX34QfXfOobfDfE/ki5jkjlZi4ojlZ1HkUgtNM3YWl2abDHv26VIQCOBvH7AarHQ8FbYQGSOk3eMEpdo0414jvzgJZyk7UuC6y+xn5IdjnAqZr8YpYZp4H/8Qva6z9pF0Mjr8aAo1ffuPMB+zLwQwNOY/VUVjEd7c7qZY34w61q+1tap+oRX8dmIzdF9jmbXHaUe9q5qq6fZVob1LQASrdjC+BgcmxlpKOpJ9Rz8pQu3fy4y8jezaoLcMbiUaSPEU1z5DUxWxetjw+2XhXbQo+kZYRaYk/+YwQcBOEn4UjnffypPCE8YqDTT7T7NsPhaLv3R9acCnLLDmsHER4t+qwmOIxGSYLruNhKxt6WFUgOfQ5lAstiHAQlkk2701zL8PBMK3vbj4h1LrI4/pxVtqoJtPFUHib+ojXrzsIvn9MNY5FjLxhGVv3FL+ibyTF/U8Et33dhNYILb9ttiOSxqvYdqr1JYpR3sgDFb4jG1QyYPR4b+VjWDJKaGghmkJv6UEhHDYlyzKR7J9c5KXJenqdI+KRLj12sryZpfxjJ3Uc4mcxnSwJTu4b61YHIp5U5RdfqwafPcTmNUojxAxeukoW3H5gB5XcIIydDdW9vHg8upUHltr9gTL8V3v6zL4jxkUnld/KYQuBP2GPpQ2KphQBnv0fBAHv5jYdib5WAwNniEblUAn4d2O8zI/LzWeeheClbDY/IePSBe8bRZtfiQEHZWjZwGDrwCDlFuxgzCMzNh+HZbC7BucD7PeQMrBmzAygk4Lpbq4zskowGNVl3Sx1AQyvPsQYYZkbDQQNAp4dwufYDAaR98bhgn2mg2Xm0RqQ4b7bB+JpDJ0dTPV+ik955YSPWIWh9og/2EuweKUjYShVSCE8AUzwBRc9buJRe2Snsfit4kDenlNhWi+2aWedOA5zRJzf6U6Jq+if5Fg2o9aCX5yIHLpr6S7xPAV1Y4cf1wwu9U922j2sSdUmH4soqzSIfuVSgw2ICD25S7VCv0vEM1ZUWnpQAGxKLSW32fY2XPL5Ra9QcSOMdv3X8sq3wRbeOtFTayCCIrtWbs7kXx3397BE2nUXsBqUQydlHQG/coc2x5tr6MjKjjSzmh9fUxDt+b218mNmRCg2XICPRJed5vu4cMWQqTrHIqR25eaFRXo4RYHdpl7/j5kt2XSrgHuJBTsEniixtIz49nALe44ZV0sR1m8ZoA0CHwSnpTENFYu44jrZWxVa/2zc4bKDkytGeZColFh0BMISdrS2SyElf4YmDZECjHu+oOlA2DO8P8acS60kr6h0MakRbgeMU8eo5z6RWDJeQtmRl9VtlmxMjS43agfpRFMWIIF1PpWRHLDYLL8VeZEEH3wsJ3mSVIP4jTeKvNn1DOaVa37+39zNW7ySYIKlxeUsWt/BAiQlRVEie4/j8jrdgSMKdKaTWZRHOplWBRczVG+EQOK/CIUiPl834gwpZ7G9E7fB1jsWWt3o+CKw3Pr/0ZXuL2wIezxvtz0cAQNTpy5PoSqwzhXBeR+jtgvYL57iq1PyP5k+358Vfq9mV5rqvbMd+hVochd1Z0icp8hLZT6NSM+OYxcqp9GwY2JIOvNvLHFZ5hwVbIpqSrkKkke47jzKb9cfw5rBK9besFAr854AeVSTgG8jI+uBW95w0871sggR1VIgissLfgtbQq8QRp280LnSd3GMQrJhE5oX/PPzLblP8vB7TJpzTu1QwGrraI92JtZ4uUSVjRgpS+a3Oi38fcQeVOsmGarM9zT3L7fFOAe7n2pcIk60XqQeCl7w9YojtKGmWCR9pAfQWLD3JwznRlvJRcmsGHPhmRVPzzwFQRInWiVsvZLwNwXg9tWAdBSSkGap6K+cOvSsf0gmaBZaXDbzWNIDalpk8VlhPsNes0z2qCjz2ncXNSJs1EIQzwp3xkWojymEdLxpAo2hIhjFNFT5wvvsaLlKXuolXeW8z1s4VKhmYHmFk8/2a+xhn10quQeoH+4qsFm03kgT4a9yuJ5bVxKo4aF8c0/njjy0dhKtEq7DUjrRGmIQb7fsWvmXLV0SPrbqDit+ukVblZ3lVg/KxP9M2IcZbZAYzjZ3OLoTAPPG1mda2fyd/ZOfvBOjVWvA+YSljkptAQe7utN2fZp5UuSK0mbNGGEbrzoo1psmXKQbdy9JRwf6PMF5njenu818aHhmyLir84yzQGOIF2vwzTQ62PYIQyy0Xn18A3rhfOYg38Zl1ZFGX9alnL+PYTAFr/Sk7tYKL8ijhp/F+W0KwSb2pTFZXn4QJPO+Jy4amFsniDMHz2KeucaaDgTLDHzCsOCiKHCyGXbn4H/HX31VjIdOfe+mSpYyFfEp64Lgr/fWjk2lIsIJUy0spFi0+EAoozQ95gS0+EUBezqICdWT9fkYIZd4fjtZIybC7+NpK8A0h+euWM3pCwV0bugJM2zR7Y79LkMqBkPW6dnpnHBRx9uj2+w9WinV4QAGhk2Vfcb8nYvmZMcZWFwCpv1RUHFeTKk3CJpJltYShazrA8/sVyRoMBFXxY3YRCi9hiUxAtdF8HQPSHqyBsKCLVsgivZPMZPyvsJaKknOu722RUF+6Oly/LPfmF9j/GuvrM9hNS6j/paAaqPjXJjgcsqvb5xNAGwlSKAqimrrZYNNVU/LS8XA0j/8XuNDCr4UYflO+ux6zRoSYWVJR6hgmGvqZMy5RJ5A6DVV+PmYyez/ZY6fTaXedhXm9CF3vDjnwZ2Ne5ovk2Q/c0Yy1GirWsnHxazgWRX1+OxTRzV1vibl/P25TjD0T+ZTMNjCeFvs8m5seyUxvBkXVEjHCTvb2xiwTT6hlAIDH3Hjn50XFszZx3iC12tVvprE/Tw36AcERp3sO9Rwmvy/xnFZcnOPPLDTW+4qKvJwn3aCJP1Stm9QYRMsOaKbB4PnZnti9mlaeOgavqegRRBiUate+98/nGvpksJ2BZfY+wJIhfXbqkM5MXi4bJAOyXofc8lxHWc5/XWYEoJecKBneKPVNJ3TlcKxymyTGsFGyEfeHXj1rm+DHfva/0B8yt/ulIk2cF4P/81bJxYI2p9cwdqVd6ZJfrD5l/dXVAZcNj9e6D88K+1CU3qXsW81JAFJGQiT2f/drj12uGGJSYp9GZZg2Dfu2GcxxM6OHvom2xMmKRONvpiU8y9IfLyh7imx6lsd73+PZF/g9lJsr53qCTIPWgbL5xSzk4V20AoaSfASH/JuNY6AEpENHhmlRpPzTA9ZDD9/ThNghgYVNNkJqqfx2xUVFk4kdxJ4JpiNJv4G2MJjsTG0i77uMox4lovHIydvVeIzOcRFZZpAyow0/vdtjbeb/SlOq5QSRKehE8HxIVQYWMaN5uv2vS0Fehwcb+scju3/925FYw015JW9vDF5brPF8bzsu94OxfNaYaRFjpQuPEIkPLumhgMmePf2D6VHrHz5LxuLeUpnLam0sUdO49lY/jvoJB0rN37TxXdyTK4M7H6e4wo55iyilenNbVNMqH+ux+CJ0vIk/fo4Y7GZQRW7ztd3lnlyrIwEqbKcE2rMcYVYjBs0r4FnPr0d9Rdm8808LGqCnmXc8v58ROmkrC1bIIpq3RfRBY7LMz5lkcHv1tcM13a2+OY2hZfrnGLuPkMweNCnBcYmY4naIZ+i9Hd8opXsOFTinycaeRumyXU5qPVjTLCGkg1Ib1yCbAJyPaY70ubz+zVVDAKciJ2TZDnRSE9sOtbB0/XknKsXLXj2eYZM7D9zQAmisgni9F9FPvoBrLHs/h4NPEJoQ8UvYAFmSzF36PRbnapdZp55MWq+BoHTxOROBk8UsUvM1bV+O2Rh1Mr3q42mYniIH8KitRco5sL/v5o01x6vlP7ydTrlUiQyZeM2/ARRUKpg6XWOWa/hJILvydehgbEK7lCd5DB9NZjNFEyxF0+4901ClboDDFPRFDajFJzDmKZ+tNyI2MgEItJe2e8Z/TfXEtqQ6rZA2GjVPq/xD44pslJhDcBRYJDuNHXaXbTqDSrdmRaBjSbt9Ez8KE2i0U5KRXyEHM5w6wUPBYGqgzx1YxDKcON0+OcugPJzuL+WI+a39eVqtNmW9l9tMwmJBCb2GhpdMikBYDesguXk8H92YXElM0ATE8UnTpP6WzI1CgYNzRRZC9tr1xQIQJdhy3gRpQe88sJluBA6iNo78OD32WXD8nWn+QpbjQAF/qM+tYHGMDZIXTlLNuisoS/e5IxMGlaSlpgbc2jk+GdT/EeEIwtawPdSjwAxX9Oxc1GFbrlcnkRAzonzkeXNX9zimNzf9gE2535nFrTKlcFhUbDzWytcM77HPjUY0o0V5/bQpA6q8rwSapE3BOtm8/GUZMBCWJEmH0i1Ioy8MZG+0MdnXYGBqBS8lOiywKg9GL0EbyEVw3Owwe5YNvyumDPfCW9S5Qu3721+2r78SyvSt7QRs+CGQwmOLk53XqbRcmI4RjZMzZ3UgKVwB7OSaGn1aklgBvoScopILlkG++/vYHyM6kDFj39tIMtppOMp5s/vz2cDQmgw7C++FnjVj6PJvg5MlslHYnzCHqnh4l8fEGJX2zerKftigwKnk5WhozIk8U0VbMPuMddHXdyBYTRiVV0NHcSB8/j2HVZb9Vvq2iReuE8STJrfek1Iw7iT0ddzwAjZjV81aksUskxQYFENqbZ+2Eh9DcqwQ8x2gjSq3wgF1+W4vzgdUGIiEHkOX90dPBKSVfXV4GBtW88E+QXi+nc1xd7DO7C0j7eln/MViqtrh9+BJ76pfG+WIc2e80gsPKrKNaiiW+7fLqFr0RBybU9VwNIsjFw1UOAibtJcxXNZzpS0884SZ0Uu9wkQ/UCUPlVT2UFmvmgFOvWqFhpNfonoXFYxYYTCyvJB6M5i7fleVGLjg1S/4b91d2APgBIMupjBnUMt4YxDVtnhHgt83Fv2ab/VpNHMbFULfLLtavV+lJdb1g0qxVcKLJr87+rHugAMGHhyPu+xXALswfw4a1VbB4s8sMZA5WTicF+7t0G2HZ9wASJ65tymICqmKGY4V00RsXiwch9tWvFRBoE7I0G+G/BsWtnJC53LMLVd/Mk9jyqWlhSQ+7S1JZFZFoL3K/upKjK95Lfz4TVKQ6JD4uMkc/WPwiMvWa7OqUDaAIL12AIGemlxknc/BRMEUja9e9QXDqYCMMosXjyTR33ChsPpZav8v2qeYXSmFIbtNO+IQBmgu+02Pt7Zttnq9WFLfHEJC9n8g6guJLTSm1PEfel/WvUJS2of5H3UWf1aLkyzyPgy1tv5nEtebYWuCHWc3Nivt9VDn7xABMrymSBeq4nm3CYw8dyvRUhUFY0ww8vFSi96Bt0AxjVGs8bQBuKi6PNOhb+CZDQT4U4RBGJ5bAIdxl16z2o/1Rwmo/RaG5Z5FsVBVyM7Gbp71VOB4ee23LK0bZrVo1Y6C011H2qlwIM/ZhBC7H90Tn6JhyERapnsrcQWYFzFv3/VCqhkyx8lK8q2TR/3VA4yzqfUGyfuJHIccJ/av2SoXJ6ZbXe9tgcSJBuLoDZ2JAhXUTaoiqfPeIo0WJvDYsQadaBmnrWqcuiZh7Fie6N/T/DCM6kt4KBsAEI9KqRF3A1Pa8fBYIxlesnjmhFSGLM3Je+xMFF5c7AfAUhHRsv+tfTKEKqIa3uAz1ObJ9AzevQnm5xAqapCBS/AJ/4FkNUVzAWTSSFp/gOMkecSghd+3tqnyjl3uLnyt0Ihyk9/bpcR63RyZ6DdleaE0fSmxFG1t52JVZuyuNlkml69B7N0OE3Fvmkzwpo4N6jlPipwwvCdt3LVM1VZS4OWr9ngSTK1IAmvN7eVZzHnkXdU0LbLdNiTaAXsx1DxHrgzUFOMM1mGhsDu4oULPbu2dawzK6EDPW8J3fdxa12BxxuyFhnrsCRuK1jnRiBMH2JDGOgei2AT97bLAe69kuaeIEYsWy1V0tMwb3XpVSWRk5Rb5CVi07I5w918DjCGPX7KpvymwLljkkp/zCDsLUrqLyoMNa7U9LxXf+gVXISnTQjbgbcaqgeZ8aXiSRqwv7v8xyFyogjvD8yGJxxa4rPNuN2on8FNJZKCb6Te9RfCl90IvKZ1X0S+/kbh9Fs9QvK+xUas47OtNiwhl5KErm/itDV3HjeHNbC64/VVi5CBC0VKts8lEulLAe5My3Kpz09dVCQ8jaLVitijaKEYSxXyA80uYlvQBAihrcOii4O+FsIhn/NQKxrcXF0bl7CO5nZh/62tvssORccSkO1tk/NoV4vtoDAy1nNx+Nq9i+GLCQEihkD3/mdyJFaPeXrmxbpDRGw7jJZ9y6tGWqsRqWA7JMYg6LrYLT4QYi81HVI3TUOucyBm11mnBMM0wI94PR0CxOQ8ikNNHGxHMpLwy1rxVl8jECirPGabcfvH+YvJrC7yztMIrGkdwQJqM97FPeDycLjBynn2VE2NuwpDASD2LCoItAQeCh9SHdkJgttgtk9mVPBXsvZCvnEFHB/eeLsj3DdhqKHzc5+tldyl5lKdEO+HFn2M+Y0sA8AxZiEpGr121xS76N1qDj3c2NDj/lwrGf5Ntte1ClVFGsRC1qRyxU8JYTiXb8nBIPf9PLsSQW6ZOw9dOvb3IuoHCOiV6+Srv7qUThm5O+Df08HQNtct/w5//AjlETVneSHzug2vi+AufNGz7jgsnhF1q7N5yzGcyPdKIjq30VqlA72Fddi0Is1ZypcSMq+EQ2oFqN2+PLwNBdNahMiMSBWWNyJaam9cByLXjvwnchAQ2UDNBIQ3N0hFe1jSp7zWt5KQ9vHCSwSiy4u6CzOZ2JgnFXGiLmnNxNRdLl1Vjy7yl4UKwJ5/m3+Vm1wd0CIzugdyemHOqSS96V0gKUmOpRyBEZ9xdYqZu2xvNPON1ZygJ7MitQclbayt8KOctyw9KpUR8n7DTE9Jjqp+fCew80YtQCgZQraz/dE1taK0RaHMlK8d9NwC4XohNvc0P4zwjlQxh4uOqVsaOhpBn6lxWKhkrAS7Jm55JxhSO7XkZ7rp9iFP7Noy+ZdwWktOJ1GFXwjchNHjzyPHz35xecGVDrOo+WPh956FBsiBCoUvjRc/cxmqVq9nknr5LRamKfeKaZo+pYRqL5vGMbF2k8P5QUnQSKgOWHqpxs7RP6wYyt9pB7UHHEf3ygwOrSaW5PL3Rq3G69LUKgmFI1M5zhIC37fhdRi+Lntz4pEtvrtqMl17DoW6EobqT9Rl2HdDZLs3kQyotLgFkp4Y3hz1pan+NV8QA+zPSZuu1OVmHygqEffOmKGlItdqbhuE/R5leYUF6JPlaSscRvBmmomkAIH4jDzzrmGZk0no+tYXCqGwBNz8XFRgAl1/6yt9zS/rsaOmj/GGPM7R1tz//Z00SIuY5B2dgQ2a2izLDQLgBpOc1V4HpDWaDonnBF0OqnWHcHeOg5NwEzYvO3ouYRUjFLz24/WuUngomwCEEnRtteu3TJA/qs/s1EdWsJxd8gYFFYJoDXtYIzHRx7x/zFunqmyQpXcZabGEo2XbmCLs5LOncQFx95OSJvmoIgI3FLwYZhs3E0wCyBS28VkGO6fQD7WN+LsT1AKIbbDXbN5lps6p3EsHif7R/qVCwFm1B3B8/Sd1OBscWLw3fm0jpCSdUiNYowuRSEMW80khlelt8XsNTOXImNnrsDnAodwI58ifZR6OGjIWR4N+zbbZ+cFSUWvM45vbolGlnNoClW9Cyqw+MnfnQD63i8L81mTqH0dIcL8mSk3Pk7C008zeM9HXDqeYdg9RdDC3t7JtRdwWfX30sZJerDYOZO8PX03EE8kDXACwTqu7F+C9WFoQuTpaN0/7T4t8MwNsrWJ5PxMwin/Z9M3jMXm1RHxPq1MLCEAvdxaE7oOkt0vSCJUsbj3lhw4sdDEFdnZ0pNMoMwpuUvPxVVyc7eMEsLq0EO47fzOIKz3UQu0fKTK5a/TDhcF4B/uwmGBEzj5nGrJE3m3BsHIF1igdNFRU2V8AX680il4JzLjxe+X/yOm761XBmLy6cRKgG1iwDchcFM2c38afkvRKzaopC9l+ldR1lT1WZ8xrRs2Tmqn1ot/VFuleTCuHPwN2HyZKCQaqojoO5sj3Qua9BH9s7+jt7gn3WnNfzBbntHd9T6s7q5PAEkqIb+OpXbYOxhTpAdR0xTdD2YO0bFPNL1b9iq9U0o7H1B1bYKco+AdaOmjQjF6OpmQTGKsToQC9cFwzakk4kNAbr0qqLIC2nB7/iQm2Aw0m9dQv3yR3t8iCKvFUUZceCA+tstJIA/KUZfY3Gr1zQl8xzYG0q2XHdVCjgXT43aF+MR6ejH9t+FPOZap+XKA20bHoFphWJx4aKRCWyCUYyGjYXOzwCiScNMXtj5LjqNHdkb9NYIaYyqwaK5fpreu7zDNotklo9HCj7q7jUWCXUipaeQ8iWKoy7sKKOD91caswoXkxi1RFCO+jJCqOmvSLg4ypOfudmiMq5ENSSHQUdciPL3CORljAieF7sjbB3o336A5NX6S+JBcW/Uzk+a0qTikJaHKFCZlg20KNSB6DR5YV1dIwT81lmhl7da4uR0A/HTqJvGAAEh42BMmRnfTSBeM7mF7KI71241iBYFpIvRW4EaYmEHFh2MvleMMnj/6VosYof08i7TGUJafXIiHrO0qK2K18MZxhE66lbvqrRBveY/p0yJQDBiSNMzwhijzWVljZIz+nzWqY0x5yIDm0OgYC2z4naHEnJIOdKby35cPujxwXj87y2crVHxhpTGNeMb6SI1O4RB4SEYKVwGogXUqGc9REuzYZF5VI/cRQyGX3piRJqk5c0WnuhesLFXIRocT6/HWScPEGQM7v/9lPQW+augSuhHXrEDrI23e2gL2Fmon4buwMtym20NGCMi1I3DA7XAOK981SehoTIeR3iWXm0n0cplfS4o4wz8cl52nlst8H3LlISXLTTCM+HGBkNoYtEycIiQq1ZjWM0dzQqq1DVG6Wi5mB/DoZ6hXUCVqWDZJ5HTJtOqimu0n9ztVKfnSjHZJYR69NiWjaeFkXeETH9rhtM3hFjJavx7Pu76H7hYoWCjmFd69u3NEPsZXy2Ot54KKR5Zp4oQI2HaVpQCYCLqf7E0eminJ4oWPztwdV3VCyflNz42Gv3llozT/XQJXNIE7dUuBZiddG3bguNtZ6XolsbcrjYY0HTF6CHHHURCIOEwHB13kcgVfZtLag6YDrKP9Yjmp2ekea3E+p7oZ/mN9mgJwlnZ0IodS9A7EiqfHDWMdC4jc/fb6oEhDlZ0FcosURrQu48OoincJO8JDntq0ODLAKvUCOTKHlG58j/dfp1ds/n6rjrFfenXdP4qCUzpCOiN7TEHqR1Aa7B1rtoXW3yflgOPnRouOY6U8dsn3BTG0RNd8ZkKB6uZnviyfUFuYo6hmuy5mUsyHPA6+BxjqjL+AsCIGza4B1flTMWTJgsQo4VlKxzezKyFUi9Vd3qIG0V6cyPS0h7kSWFcu1S+K9nu2yFI1IBIvm3srA6hWJH9vD83MAhMC+kkguJllhVF23VJKTsC6VotZ4+AvFV+fJtvo96mpbvmQhSP7Rx+E1jgM9zENQZb2At2s6eJa9/0cSJE04FS2EqYjJrw7jfAYM4/KIliD6szpoZKmc7gDRZ8j+axoryIXPYQ1JyY6KaUHMMjeOgb5QKxWOxgkRfjlRmOhyOGsLlyCob5cF7VWr1+JzvuolbWXk5mwIHoDRqmHWTSRnhZPtpARO2B9sqEhW77XWB6g8j1NtUdHM37xWTkRtGh9Rgljwhvrc/RSx1287WbNAW/4LB81gTzuklRUSuKk3Jo6K8rlxt2Tru9b2VwuoBWcucosyHeVQqYwTpXG1M4YYMobf5a8fg4YeePC/opgPU61pfsGujVZgoTXwW7IMZXf+5iujWoyHPF+hdi4/5NA80QfIVmBmpuPmJG8oNlTz745av7Q/gWi/owVJCrzsdZb2SL9i6O9hj0+7oOwtL0i7YU2Jbuq2sc4UiBcIVNEcizO+xsZFsczFUnMIu026hArIkz/OTLzYRFxw3sp7q1frVs+9oddw0sGPjHGOC4dh7Kbr6WyksBe/341U8KZR2Hg1ToQZPTOJKsmUwzihaCstAnnpwuhxPmzmwD86EWgmivz9T5rhERAkEkjTLy3bzYE3kNlZrEJCrXEHOYliuCOiJuKxQRG75RQffA9lYtk0uRZ8JuC+9bFM+MsBvgyzKq2zJgyALBWptMF/uX5plQn5DEW/zJBVX6mYeFP/OtOQNUUCcgImACbQIIGizHUGYuvsotG54eMBvXalnFKIdA15cCNgrpnAmsAeiDpl+7wQMTgfCXHeTMZ4d6IX9X9u6yQLevDU+QQWNb6x9IMZoHnZglJd8ip+6l3dEY3ionB3UWDdaNdLTGIZt31AlJLu4wqsaBwuwm4tmdrxaxGsCwdzergXG33RVOMgZ4PPCgLljTq7a5VgYlRyG1nv7VlY0MxPXu9OrG+A7/4go66ahIOoT9NcUWC1Odaa6ZYA+Dx4vvcuxO1lpUJIUGAfhwfbsha3Y6An4joQyOoz43BKeGMrrgWAjBeInS7KmsCIqQVvwvR2weiB+XZVQa3roOWlQsGIHg56W/PBfBOF/KcgsfQmWk+PQTYfLWOVsvaKDneGhJ31YZM3/gA4S3wekpTzCCf1GHkdb0B/UTuL6MHZPDDwkYreiI+x9dtmTeEaDxekRtutbZRMEhk9FqdVN8j/NUaB5RzY9tEF3RaQMgRaRvqJREHAhshIIUkmk06DPDBE8Rf8b7Im2hF+hPDWFydj+DRCmOWbgDQKIKAcdFwCxClkbuu8UDWaRNEocw4qpf23Ul+zp3GMHodv6wOXuJ0dC/VxEqrwNyTbbMV8XzO0gvpR/m4LeTC0Ipg9koEpNbiD90reboeP6Yu797QQXMsoxa0pnQc0dyc1HKhdv27XsVxtSUc1BR1quBOD5HqwDDHllBKZbsrgK1QkqK22kNyQ0DgXfuUbNYlTo7jc2EK7i6nzx3aSBFg28s8j1NIIhu5riAXZ6LlwzPqXne3hwUmdhGt1fWrFb0eOZC8rTHoA8f/eAQfrNM+HOtTNLsntA4a2Kn3W1+z7YosKW5Q7c+1AGq8AHrwvgLHjZW0WAZUW5OGrrpCuMxdwI5uOPX5uHajXIZstrm1KYwQX9dFWfrEAhrI6DcwTjoDrExz6yOX9SeK/MP1fIN4YIz3RbHct8G48T4Vcz8/XMR01SgNq3vcx/8V2XiLDNoVsYQd3zf5AOyBmkmA7sn7BFuWm3Rh8AefyfTKaGD5w9Utohwg+rhMSs6DvtlHxpjaV0cFXtpKrznQQxxE6vSxl0Y6caCgRusLZC/KkSlO5sAP8XNm3Fuh1rILStJpRaBWTxeunQwWWp/x0eWdnmsxfNAOGI/F/lMggxmSP3cn7ZRa7RTtgqcG/l6DTlv6iFB+WsUxU68ZwEndzQWjD5q+Obd2iHPkA43CeWc1YGk3NFGdAGLjsnM0jhkUbXT6+CBv6MEdzgm+PB7cZIoSebBT+eCUyJ1XzfK+PlX1RvJVMrPRzagpsmHubyV0ds4qN74tfFO6kn92Oz6NhQdMvFn0EhGXPkx02AL7rtITrFhgV1Ok7HxqbEREB9gmCABqgg8G537chkYSveIpeLYe60d2ufE7WUa0ZLfXT49ikQAnyHMWqSSQexoLtQM7H0qIupI9578+q3+gc+InF6gjEK23j/ojjYM13vFScU43pJuB361JVV0JM1Glheb2YMBJNc2wtbHHROU0WIX2cZnm7M7tC4sN+C8/RzQIDHCEyEPkwLKRkTPpuJhClEhyC+LWmfDypJ/UycMQ0HbOR8vRH2xMBP6cZwaJwbox3pR4JhG5AuNMxIUxuPVqgD9ha1ti0zoCb9Xootjm0d34aOlq+uN9CszkjxJTjdM64Fst+C38ZnOWZNuQu1wZUfL77/RWdQN7he3lJbnzGWc8QbQPXVrTWSUC4BKMN7daMKEIIgs/2kfNSpi/Q3IMmOCMBpxdBFIfq8F8BTq4++sxxkb/BROxDjl3U+v7UF1KUBeLAVonFq2l0zIAZP4sWBxNCX8ydI9kZNKVNkwXQMTv2FbPoY8YDqreV6Iq8aSg9zPHyuRdl+4UyeQpVXC9pBGl9yn7XJnJi8VsJ6+Zav28rbiFxG/dsJzq6dRtMVZGgOaAVFCntR5HCLEdTSD3TCizW9IbyZG0v/TAqCLVihI5+NdZR06tzgIGFSZCpj7TiyqAv0tZ5LxhJ6TqyvmCME3dW7ZfMti8Ui03e+o0hom9pUeeRzg3VVdtzz1DvVNaMcDRDDaFPROYbcxXTuCwsmTQWykUbGQzhBtt3saoSQPk/IM/KrFspYd+x0kGlL0TLQcLK8WdkJwG52UVytGu8BQKBVWJj7cm1SvDNnsRSbxpQA6JjU02GfJ/LblPxR5uNJGm0dnePvCyRwbc2qVPLy78BD3Tlqn/kHBZKwj2aaon3dZP/wMAC3EoLLC0gpKoW1g9PdmES5AnvwagmqfL/ykmugGhKkPqZ8/1FreuW/B+wj13+KWRyYJCsGUrn9YPKLfT+u4380/3SCwMngP+JdL99PPoxgB65OUi6/FVJdonbEx4/bRw3jYVA4EVVpdpt6R9Ac9j7Ky3S/nC/6O2acQ8C0glRj2GHk7rYwpzq/khF9vKOPYCGEv6uTgWYZvkKQVQJh1N8AE82CwXZKfP84wv3FtQ9JjMVGD/CUrVkuTbFA0S68arCjhIipj7ue2pptTKGE10fq18GCc/elKefwJ2P6byrWWAYsYJfPmgQx7HWjFQOKrGjy3dZxRRj2ztB+Esr3B21+X1c0f9HtopI91BeuCxFifcnUP8AfiyB+xFPV3hj3d67pxyShSmAkQihtrVODePdCC2FITb6XMU5RWfBhKkR75r4x9ALH+R8vNmW8f5TwPkFZwbOB91ZxYLLS9T6wClMJhl08knIzu8DR4opkO+N6TPVYbbIArmov5cHxVqYG2CSUtJU8P5AfqW/t/QsWrN08VSEVRV+uuppE37GJuMLE9oMkEYv4Xq2YHLCEYdKhhtJREHmomRlLmjmNIUXZwhK4uzZo6yTRWarNvjcfmDBZzNLp9GvrDuVWvaPIpeXXMEDr3BHW0gXr0OPpfvoes1MApwGbExZORAtrmdHMbVhy56FwULnMhg/19Mgwxsl01oq9vJ9s5FF3xnlk7hEu7HOr8DRicpo9tQ8qo56c8eO95qIJORN5nzK4V8ZQMc9hZVqEexprIJgCl/6w1nyB+XlQQXonX14FIH/bdbdQlJNE1W77hrjeOnL9/5i9x0Nbt/cxoilJSz6n6JluyPWYY3nV/KgHPdRYBJ79oaXFWnxxkY8yaMLxzfjPdydrSkkT1P9sh78zQ0MMys92ywP2BHY4jvFOtds9ysrveALp516JhaZeh0542UpYxsbnY2gbW/SqmSVUBSqxU7iEmVc56nT1KuZJvtH5G3hJgBOY0qBPveg+Xq0/3x3sWGmrDxF3jLE63eVqlxUCcUKYvvSaX903xNdmaYTEEklcFtMMcnJCV6mjxp1eH0DvYO8hFA1mhyiHbMx51zvTLz/dYZ0YDpcPyG1yPgO3qgoYs/JECSA7gRERbdj6Eg9hFuao/u1Wh/KlujBnK+SkOSRgOzY24eoD/i4R1hYsX1zTEvkU69+5t6qtyIAM6hD4iAsOgaEn1P/aoukSAOzfHHoBqFNbZKL2ioIdgXJyJWc/S6LY1M9rfhw6qmO1aIul+InPg0OrPwj/pQ26cf5myqGoLX+5XGLeXAlZ8v2oxydtsPqYfDeEDIsiGdgTEvYRzPbjNH2g/MTuv/V85+xePjnkoUigzWZduBFtKYr4uF2ia0xvZTS84c5wIfBNCbjlL7QzHEoec6/HnwhLAGcFXD9SziwFLmBUySjc4EZjmc6r50BleKt2l0gTRzQmWBYj3/hiqRNAKuVgxueTk3c1IL2dpCk+DWzwC3+0jIkBKR9I99IASip9gtqeTW0Oj9qCXlqFIsUSaVd3f7Kkoe4vpQ+FcpKkfqdJBxwVd/qa5cxRLkM7MhbUoStvGHMn+3wDZFaIPR55cfele3d2lHtoHPiHhb6T0oovUaWR1Tq7spPh1490rwjD9RKAwL5gOb/lUBE5ruxkkG11FQ3BQ4RDOqJTRXHLGX2Eyrot4rjCxsQASi/9uJ7zaocfVCZjMDehG6BDJTg31xbfRRZweL1CCxzRuyliaxWYjXMJ9oLhkdG2O2kqbtR/P0COt5BjuFuICgkND9XG5WXuXdeZessGa96Zwerl8X0LTakMy+77Q7LbZiLeE+NVsRt9Ld8RkZvaotAT8tdrq8niRIPGXfB5wX30tVEx5aydOxrQ9m43NIKFLTZAVZ0npKpkNAAQxPuY3DiCga960cnYEO+3rhgHufLGEdj5oW9k+yy4bb76fGU9nwF14Gufp2PQV7fpXmoPG4+kYEfw7WsPz4rPvV+C+5kUYQwplDFz2+jaLUvbaP2wo9BlVJtKP16geHsGx5cGLwmfOc38fuC+pK3+9TUKbtyqLOVqJevmfCJNmUXBwFpyjH0PpNAHMhc4v+euNs0goQfDtOe3brLegcXNYxbAHvMfo0hvYInCdAFFHPiZ/T6t16rvkisn4KPxdErvmsPL3vPaZjXxK744lKDsQ/eojF9G4mHVruDgnd3UxvoW2lwYV/iGvdMo+1dtMN6ixP0Dr+dG/LkK2sjKCy82Nlymd2L59f+/epFZ5fLFGz40ibt1wM49XfdJwpfnB7FE5Hcp65+U9DvGM2lYpujz043Vt1HufjkeCOoFxr1L1MlqHcxK9hU+j0DjuzQz5WfCIQCBs11MNdfnZsKPluAcqSA2IR+sm65KaPokNrzA1Go+NUIJCKcVyoOpESd1HirVKhDK8Oocqdt1z+nS/EEhPfy1Q7Os2kAw5nce0bf5ny8CwZ64GRZs+Pnf1Rliw7jJf2Y7bxwWU1FJ0jHNSqhI9sPqNOcEzZlJiijXFnQmX5HExaLGWaxEq2aV2x/SrQ20JgfVBGd30dJBsGAwDAHEoUiwFlxLYSV8b7FzyMmFT6+u2TpJGehfa+l5V9RkE5wonHsLLm+SYzw7ThVygYPXAA/+f/CeMTCuqpTnu9LecCPRHM3PGhL+/68NlcZEKV19XYF1HbUDd2nJqRL1/k+ozLE1Ud8+OGRbcqW9rxScEAJ3dx1tj7i+KrCZQFEpt+moidt9nJURyCGKL5Z+T6pUwEk/nh3tviYVUXazdXb5HPh/fI+Ov7URqmPmL8bWDrMncUcZQoItYuDwKP21Dvvk1injvwDuktbnA9Ijc/MydM+EDdjNLhKJWWNrUxpck5sWfZ8iNBS92gpx44g8JhWg6FcJiXitJM34aXGIPBtG9Ts95RKL/gVs8VT6cpQVkz+uJ1teokeh+01hNuESd5ABJUGFMV5zTQxkbEi+VbqeSesJhG7wsWBOoE1cYYBmcHr/TOrfFYk5rQcYlQjrkDBUE2CLLJRPARPd0p8MrdrV2sdNG945ojjD1R+z4bUnJs4iReufUQ5FLirpCmuHIAonPU/QClTB01l7dO+9SCvuKwGi2UaNIMxA0zajdpEpWCBnlqahgNWrHfWaBfh/UiK5/jKI8zbke3hdEQw5DGuin7M1UTc34+m4jfxHyrUEwU2txeMez+5mxHbr8AF01WSUvFRv1lt/gwN43jPCNSLeca1h/EUfkxs4j19eWYjcjr41G4GFb8FVylbv1v5nMkfzMWENV0uVvRYzkeWMUClACTE7jl23cWUx1gScCRmoP8Q2+fPtuC3bggfhHfcraYu9XOkom3ZwgTC600EOa2cTFRlWOAWN3wg0zEXjJ99J23m5Dp8aHZ7wmt/DQyWgZdSjzwFLnOhNNTwP/3yZS6pa//qiLak06iYoT5fCI2YekYYzNnzua9gskFUqg+eunHHo6AxxVaSqy/DMakQtGndsf3R3pbeH4YqlmPFKJ9UQJDT7NkMJ3snIOY4dkqmKtQV/qQp7Mg6QlR5Mbr7KOSHkuSo0TmnNXAjXmUnStX4YKCQJE18yCzVEIwT3YbSh33zLBg1No44BcEAxRFCZ9uTRJu9UK/Z3BPgx6JZyoasqgA+D4zQF5R2jIPadcC4ziwDhFMwVs9OpS12+/8Ht8wD2tEKbF98OxkhW0xntGnN50DM/ZIfWrCmQKD3k/3WGcXaJBwNyqzvkCEm42S5bmdVYbBv99/DPhpwBACJR+LipLQSMmwiDv8qmd5w1qlnIgpjf5dbbrHeV3/bi0O5oROlevrTfVFf09K2TmpXfJkT4PVU+XxnPNKgpZpWRROajWAVQjuD6xg/2gaGZ/Dy6vVsHA2qcWe0KLm1yggNjJ4PhpGohqRtuvuCvBMn6z6AY6ZgaNjSbdZZtIfBQM3NhddAHhh44+pXxc2wRlyTRdImXaiYJrtQMTLfr/7P3W7CwYQCrhQGX4iRDH8aCvEcNP84WwrNKzvE2aDRmA2KpiPKlqKWGE7t6NJwW0CtIdeBGzbnrVsf4fCTzm6uYoVKEMt1WuNCIzOKArKhP+lNgUrJmZfgJUNWB37Y+5ELkwYLWGPVRDn7NynkqjjADqOjItxOtYwWPp5n1EEGO3/mfPW5Li68t/1nGUuJwu5nA46yFmmEIzhMRlVvz4VCd7npZfgoPPRXsbs+l45Qpxz0vSO0O4HoAkqoSi79lJRwXu+VmOrshzJa4ZvtbyDkGQ3PVlDW5KoWwt4ayCGsCmtfvzbmt2xR+LmZI2h7js32C8jmZL2BVp8hlUkJx0glISUa0pOy9J+BgrC5RbFuXRzRx93CqrFt88AK1XxQGkP6/KIylHnU0zqY/9yom73rcvs2OC97KEvSuT+9mzb3echLzOgGEv6PSmBFzwY1038R7wYamgUWvreaadW5bvhdX9PhaHdyXJb8C/L7AJiecdk8BQSJ/itegJW2YWPJUzqivdhvxK3oS2EKNLhBlA19K5Sj/PKzz2GxO72TA52umFu/S3ve/4lhdOH4Z3iVqU1W/PvR59PRYgl6XlpoZdUxQU3XD6ZnzZZSslybNFisI+KlcPkVWaNKS6PjOLt1PH5ej2BIdlihIV/9g7U3NMqQpu/qqXR/6f3GUnh3UR+8spqtoWAWezYtJmkBuLKyvUG67vHWvM2b01xxGdhJDoPtoON2Y1JYjBDL/uYdh6HLiFkV8OCjFBdgOm+X6NxYOhvQcFyedCqAUYfZIkQsQ4gh8Tj2hdmGTrQwggcdHExVTMYz990QYHXqOahJ+RwBkGNF7vFF6wyMYRrE3r8Gf92Naf9GJ977XmXfpvp8W9G9uU/jvy9JBX/8ZVPSMC2Ysqx0ZL4cs48pGEWAXpGnvZQ+ilmNPgi/alXg0ITSYb+ZhdthoFeROttd/NNVTi9tvY0suMMYgzNavDv8X9X62uPH06N4SRc4bG76NP98Tovz62trcndKrwtiz54iNwgYQ0Sdxbq13bafc9hjJbrtehKeoW/Ehs6MWn+G5weC9j0Lv+qR1vDVybkmhEZjyYoXf+nAvkE+xc5CdZrD2R6l4Nmt6HEkjVPOR1hEcnOMSJ6Wc4qlEeA83BZF8zKMpuRkJDC6pDMNgs/2v64k/c8fHKsR+sURBf6dnA+sl+Pa72AKgz/DHtdRUp8enZOzhel4ZrVPtbyGh5ii/86c/8uVjZiOEilAn1x6Qg31IKNC9UeAx0i/oM1OzyFIKfNv+9iN/7bbWlapRz3wJed9UMvqjMu5TxLgVKLj6imy1wGQXEVwBEjLzErm58IWD++ZCqaWrq3zMaQrUlJ7lQ/8jEkytZJXLbSmkWtDYyP+vaGz7kDQdpXyc9Qoxu9ASkxSRCHK+N7sUD7kyARgUEVv29yWKUoolfQFW2XKbgYKQ0RpzJyvFr7tI1BMHU7LBMHXKO10NxBFa52DeHZ6MkHnZSCpnNcfFllfg29YFtj2ONx5zCct7zdvyJBDL/V5hRofJ78ITzxhj8yTMgsKq2CZpchcsqvouBRUXyaAtgjGFbzk2xR4CSQ52z4dN2xpk5gdoSQkmMS5KbJjktpwCkQOtKj/oxNI9yW05T0SQPbLIFwn/oBEDhFBg+V7Jsk4g3HmdalOFGo0tZLDuwuhEb7aJtF+DfTfhMJOYHqx8muS1YiyJM94BezqSca0UZr5T8amhOSy4xmApd3/p5JOPUwhvsn64ywzJUqEiFKsdV9qcaAmoQCHvmvLgKCFRPQ10I1bW6Gist1LdKOmvlaYgrbtWjPIY0g/vNB3Irrxw5JqaWcRkdTTRqogPT+2IK8c5mIVObBYUSNc6cqmuKB4cmNa5djsXTCVN9LSI5THrpwF1kKnek4pMPHuQOrJmsh4C0QX5x+uYs88qW69Sch2knOkaPuFm5Qa+noFljhulk4gnzVxIAjl1n+DTocSMAcpLM5W1abHiSrBDvYY3kjVObYF/hK9ZdmIQzAOZSiaeXZaML+WuHciqNrDsw3zw3qlPDJNttGyBJodwxBYuccIbfGY8rNhq34TK2DtTke+h16iEASxGvTb6ojRSY2xCsQrysd/+isWNdiyUvkhHeSUMbu0o9JEYPFm2vr1yWK44e7N7HZTdcB5eRYsjG1Tz3s4RxcR/rMOxXkwbt2MhOpN3Q98llO9e8c3iU6VTvtShKnprh70YQov5hvgVE5ZNukVIHLOMT610qAfZE8pirelrmomSFxRx2fpkYWLjc+eeqWi9W3Adxn3nRyncVZVOZnDO2oimLBpF7+vvFZHogd7HV2RqLpnLj//en49Kcjx/l6AgL0Dixd0/JEU9ejgNmhZuRtV517LsmHTh2tTNj3I290L1VYtkE4YrVX+0WAzQMHH43FRns9YBBGmV7verhhG7LrOf8SZ+3bNYqqh+XvrcdebaFecAUcKRSWKxNnWNFY6FCGaDxrgIrbUWjB+1kXvFZhR6vSSFacx5wkWaSvvzUaeSfBNc/vkiwLcUGXztNSIg6Gg1ce/UICOf58ijwuiJf4D/HgRKh7BPy7+1WfJmEpDwODizu8OQQ8b+Gl8FQ9OnZdLn70WwgCWRPVosMkJdUSeNuCZ2yqzF3BK9SCGFnKYMun7xqTEHHsKqif3xJhUUPnvXc79URHFHYYwrS2arFInEU5rKAqN+ZkUxFR76HPTN82mbE4MJrXhh6U4kC4RjnpbVfp7HlQfj0kmspDc6HBRsy13TlSzg5/mvKYGZZzsozHDHKOfWkxL4BElR5ZrSmHUYsqmtfIDgdIXp/I/xki6loLUk1Ibz0NJM14Z96T0CL6oHnlfIqS7aM9LOHp7lJULD8IV1v1OlbJuOxSJ+Tnz2qnECiJGs6LI7DjPtHrfh18ZQhFZxuwkKZegdoYD+k8ZXr8vJApL+VKcO9tV0Pr8NdJ1lnpzVXTvXKCiurh8WWPOi2tgAjfTS5aOuz90QKpRZp0sEvJS95YLmEPSQrbm+8W4ztAvu8iqUoVvZgjH96Ji68LpimawXmkHpxXuSxqn5x+cg8SRqnsctEQrOSmurPMRuNmovXm2zMan0Djvvxy0sTOHWyFjLa9bCmxLWL/G0E4Im7qKxjXhW2hQTYpNy6JzocxThQH/b63/wpXOubdWD6iJhJp++Lq67UEnOkjUqfQQW7zKNiZUC2rX6aq+6fNE6oMhgcU67cuYH/nBHuPwo/i3O8wVo6t9ZXVD4F1iTDQpj8ae3/rfybT97d6yYPcgICoS9xfyYWd8C8QZuIjPPlR3F23JpamuXzvQlPdFWqVWvx++hrXes7wbvKeGpXM3lphWgai2MQ0CMXqN1Pz76jKrYUFfaWaRGy+Gu3WXgjc/gWf2bwDa6F4jVwBX95Go4sevuoOiIinHZcFWO2A8a5CxecvodxyLt2pv6X/lO1Y/o8QdnqQrsS104wwX6ocvitelTyZ/nCqxZLfCIl7Y6TvsgnkBvyBcdDOx9VpV09+Qp7+WTUb63c23NAgjHmnWii1NWX5J/hqvhPYh7Fa9KpSq5AT7YQwAijl38QGUHAkY38+CpZ72sdeYIcLvZ+jjYyVjSqhl7TXcIBVF4BuEDYCkO9q0oxsXxVOoVEt5TdeNxonNI254azGcvY6NNsyGKiLB/iWsxbRVRfpwW1C5D4K9GuDiVGgjXg4W5U0rGesr7qSQ4lGEOr/8SnvvFpJB4XYegqdOSZiqVJQM4X+ut/QW9D1GM2+z9SNJbRLHUcQrjSddTZn0HDy5UB5i//gcJ7KMlvi8fSKtlWlecSU2fH/phdzX8vSlwsmVtANW5KC45UQAJrWPX2uNlAixnHt9d0Z30o2qEADIamxtFR44YUuW9DhZbN3qjdxrixzubuUhVmXawhCeUKZBrK382JhtOXe2upT7SY0Rk7QxNp0MHlyPXbFoW8SEyiUKbZROP1bwgpI0MML/21G+H4Y4Lj5RX7k975vpJ9UZYW38CWmJ2bKZtu29Gd4blzmPC24PbM5DRBd7bP3glx4KhcAI6y+EPuYVNqd3JhnLcEVmLYSp6Lfm3G570SYYk76y7n9kpsjpN3snv84wiJmnJUVfz3nHgkGT96HEaFLsdFO64t1lAU5R7Xuur0tTMGwx/jQ/iX9pgDXSE9+47RJkzDPyrS5EZjhIZrECYiaF9IQHPN3A3aPpi9qn0dxarel9m043twl09DSE7JTTYGQenI02jIRDLW8ujHXh/o/0FRfwedr/CVmj7trrMBVaomOsRLRBGN49+1dK1ZmfLnwKMG054BavMvzUtnB6sB5zxmJ9GITyNYwFYQh9zNmZTOdqmjv6AzCLzxkzDqDZmI6+1eaAKmaN7Mtcl383XWtL1Ai3pyKquOcnamkG2ojngkB/RrsnXFU7bKpYar+l2ewKp53ArQOWuSe5p7V+9rxO1Rlyv5NnbomiISPq6F2BuEk90Khu0IsMsBcjx89yB79bsgOTBKEIgXef+tAob5Msy89m66OVZW6oLsIF+GEqT+LgmktJSrERPSFYVFGk+5TgpJBge8E0cQqHqaBn0KGgqUQngT7pj2Ugka6q4Iw8BEXYPXV3jQ/ZJc0Q4C2FxP7W55cnoeeg+DPlX4JwfbYSXMgPef88Z8oWDXI5Yy/im4JIs1Qfj79PPzEKVUWYha5TMKBzBPNlrAeR2EcoghAB5/S5HH/zlUNfaTBjFzBRuYuzwd53Mgf77TSXCEj1l4LSNdCYOW65zmsOOcjS7gT/TiHpLcKh8nfm/z7Om7nkNEwC9tuNfSbjubdVmDw1ltaaa/vzPHeb0Zv+Bew49of1iTD6ZqDJKCQAlCm4rCt9nKN883aMw7kUNtXcqLM2Ww5bwXqk5nki3i6yUULPzGpxOYR9nijw2GW74tzO4ifazpjpb4Gc9BwFT9yixRsSfCMKViqsntFxYERbhLZSD0w8T+vhfLEoNRdmfrjT4HsbmGsjEIkBqFZFNLeuEZLBTyEdenk5ga800Kh46eY+ZzAa9uRd3q36lFuNbzIH3zwTqAHyz9mAzZiNh+Fa/nsaXK+/AeDhpKARocLeB7JFcGLRrINRILpKsO2Jy3OqTJOPd/6SzaC7tXXCSZ+w/TyUszbz8sQX9ujJZ2NF7ukzYp+qLDtVYxZVdEkIay8yZHvN9OaPVgAAHcmZ6gG4FctiLiu6AGsEedyiP9CRRKKfFvyh3KmR9N/BDvB9svhvoJvf4tyFh5VpF0r69Admi2J3UXJe1ZNplTawCzl/VSSfniiYurjyfYoSFZH9lleSBzyMxaIUl38HD4Zfr+vBMd228RvlPkt+vzX1+80tDFNXTNqLxUecv7LOrAkrEbPboTF0occ6NbFdOpf5oUy52jq8NVbRLgAfW7dg904VhYWupTMhao6Nvqq1L/EpRxO3tInE1Vs6qBp2izhZM3yAlXHhNpASL7MJLbh21THjO5GLy28o5vXUrzw1zps0CIkMnynfAFph0fxO9NTsixOVdZr5ChQUB8iSVZokyiuOQbJwAvArONoEmtKOvaBqFmDaK1PLLgANkLafuN60eYwb8wVFpifZ4rFDa/9eNawLU33tyBete8K8P41jnhKUIz8R95ZrCb8Cjh1sp5f+fxCXdbGbQXCsfuTc3t2JapIKfm9S2ZpdkyeMJYXx8HRR563OikI2lf/+G2k2lnvbz7xMY/UzSIEnu67MwSatiDhyJVF8GA292pNdIJIcbnlk24UcoWVm8NZE7y6Y/2EHUuYbRkstRMBf9Isyq1w4w3YAXobGXq5I0g2gPqAsvZxOCEDvWBKcXMj7RLXKKIm8vvLqrArUNPsTEmcGn2+AJJS1bO6yWRvj5Y6rK+wG', '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'], 'server_side_tool_invocations': [{'tool_type': 'GOOGLE_SEARCH_WEB', 'id': 'nsdqfchu', 'args': {'queries': ['brief summary about otters physical characteristics habitat behavior species']}, 'thought_signature': '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', 'response': {'search_suggestions': '\\n\\n
\\n
\\n \\n \\n \\n \\n \\n
\\n \\n \\n \\n \\n \\n \\n
\\n
\\n
\\n
\n",
"\n",
- "**Lifespan & young:** [*](https://en.wikipedia.org/wiki/Otter \"Otter - Wikipedia\") They can live up to 16 years, with their diet mainly consisting of fish and sometimes frogs, birds, or shellfish, depending on the species. [*](https://www.doi.gov/blog/12-facts-about-otters-sea-otter-awareness-week \"12 Facts About Otters for Sea Otter Awareness Week | U.S. Department of the Interior\") A newborn pup needs constant attention and will stay with its mother for six months until it develops survival skills.\n",
+ "- id: `chatcmpl-xxx`\n",
+ "- model: `claude-sonnet-4-6`\n",
+ "- finish_reason: `stop`\n",
+ "- usage: `Usage(completion_tokens=158, prompt_tokens=731, total_tokens=889, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=158, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
"\n",
- "**Conservation:** [*](https://www.nationalgeographic.com/animals/mammals/facts/otters-1 \"Otters, facts and information | National Geographic\") Otters and their relatives were once hunted extensively for their fur, many to the point of near extinction, and despite regulations designed to protect them, many species remain at risk from pollution and habitat loss.\n",
+ " "
+ ],
+ "text/plain": [
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content=\"Here's a breakdown of what happened:\\n\\n1. **Tool Called:** `simple_add`\\n2. **Inputs Provided:**\\n - `a = 5` (the first operand)\\n - `b = 3` (the second operand)\\n3. **What the Tool Did:** The `simple_add` tool took the two integer inputs and added them together: **5 + 3**.\\n4. **Result Returned:** `8`\\n\\n### ✅ Answer: **5 + 3 = 8**\\n\\nThis is straightforward arithmetic — adding 5 and 3 gives you **8**. The tool simply performs the addition and returns the integer result.\", role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=158, prompt_tokens=731, total_tokens=889, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=158, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "\n",
"\n",
- "🔧 web_search({\"query\": \"otters facts overview\"})\n",
+ "**openai/gpt-5.4:**\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(**openai/gpt-5.4:**)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "Using the `simple_add` tool, 5 + 3 = **8**.\n",
"\n",
+ "Explanation: addition combines the two numbers. Starting from 5 and adding 3 more gives 8.\n",
"\n",
"\n",
"\n",
"- id: `chatcmpl-xxx`\n",
- "- model: `claude-sonnet-4-6`\n",
+ "- model: `gpt-5.4-2026-03-05`\n",
"- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=627, prompt_tokens=17556, total_tokens=18183, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=627, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=17556, image_tokens=None, video_tokens=None, cache_creation_tokens=0), server_tool_use=ServerToolUse(web_search_requests=1, tool_search_requests=None), cache_creation_input_tokens=0, cache_read_input_tokens=0)`\n",
+ "- usage: `Usage(completion_tokens=42, prompt_tokens=199, total_tokens=241, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=0, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None))`\n",
"\n",
" "
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='Here\\'s a brief overview of otters:\\n\\n**What they are:** [*](https://en.wikipedia.org/wiki/Otter \"Otter - Wikipedia\") Otters are carnivorous mammals in the subfamily Lutrinae, and all 14 extant species are semiaquatic, living in both freshwater and marine environments. [*](https://www.nationalgeographic.com/animals/mammals/facts/otters-1 \"Otters, facts and information | National Geographic\") They are found on every continent except Australia and Antarctica.\\n\\n**Physical traits:** [*](https://en.wikipedia.org/wiki/Otter \"Otter - Wikipedia\") Otters are distinguished by their long, slim bodies, powerful webbed feet for swimming, and dense fur, which keeps them warm and buoyant in water. [*](https://www.nationalgeographic.com/animals/mammals/facts/otters-1 \"Otters, facts and information | National Geographic\") They have the densest fur of any animal — as many as a million hairs per square inch in places.\\n\\n**Diet & tools:** [*](https://www.nationalgeographic.com/animals/mammals/facts/otters-1 \"Otters, facts and information | National Geographic\") All otters are expert hunters that eat fish, crustaceans, and other critters. [*](https://www.nationalgeographic.com/animals/mammals/facts/otters-1 \"Otters, facts and information | National Geographic\") Sea otters have an ingenious method to open shellfish — a sea otter will float on its back, place a rock on its chest, then smash the mollusk down on it until it breaks open.\\n\\n**Behavior:** [*](https://en.wikipedia.org/wiki/Otter \"Otter - Wikipedia\") They are playful animals, engaging in activities like sliding into water on natural slides and playing with stones. [*](https://www.nationalgeographic.com/animals/mammals/facts/otters-1 \"Otters, facts and information | National Geographic\") When it\\'s time to nap, sea otters entangle themselves in kelp so they don\\'t float away, and they sometimes intertwine their feet with another sea otter to stay together.\\n\\n**Lifespan & young:** [*](https://en.wikipedia.org/wiki/Otter \"Otter - Wikipedia\") They can live up to 16 years, with their diet mainly consisting of fish and sometimes frogs, birds, or shellfish, depending on the species. [*](https://www.doi.gov/blog/12-facts-about-otters-sea-otter-awareness-week \"12 Facts About Otters for Sea Otter Awareness Week | U.S. Department of the Interior\") A newborn pup needs constant attention and will stay with its mother for six months until it develops survival skills.\\n\\n**Conservation:** [*](https://www.nationalgeographic.com/animals/mammals/facts/otters-1 \"Otters, facts and information | National Geographic\") Otters and their relatives were once hunted extensively for their fur, many to the point of near extinction, and despite regulations designed to protect them, many species remain at risk from pollution and habitat loss.', role='assistant', tool_calls=[ChatCompletionMessageToolCall(function=Function(arguments='{\"query\": \"otters facts overview\"}', name='web_search'), id='srvtoolu_016FZfqBdBegL6dwXMvCyBHd', type='function')], function_call=None, provider_specific_fields={'web_search_results': [{'type': 'web_search_tool_result', 'tool_use_id': 'srvtoolu_016FZfqBdBegL6dwXMvCyBHd', 'content': [{'type': 'web_search_result', 'title': '12 Facts About Otters for Sea Otter Awareness Week | U.S. Department of the Interior', 'url': 'https://www.doi.gov/blog/12-facts-about-otters-sea-otter-awareness-week', 'encrypted_content': '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', 'page_age': None}, {'type': 'web_search_result', 'title': 'Otter - Wikipedia', 'url': 'https://en.wikipedia.org/wiki/Otter', 'encrypted_content': 'EqQgCioIDxgCIiQ4ODk4YTFkYy0yMTNkLTRhNmYtOTljYi03ZTBlNTUzZDc0NWISDInNO+LjQwjbLhUsHhoMCHJOc59TTp2ZH7yuIjBEvsY6dnZ9VqZNWq9MQbo7QCF/73LBjPYOQcMc00xDOO8kvc/zj6inzTQdlLOLGsYqpx/TJg0b+Rb7/z5dcmJfPuIvsq2M/E+YeF1vDsVaATvvXFL3EjjXzVGfWczM7TkARpS33NL2QKZwSry0stFJavaIX9ngnlqgZ0FbBNzDC1MhoiNQiWhiTokut/PpcU47IO90c1MZIQIFFZ8kPfNQJu7B0hT8NnhJ7YKMPWhHXbZIxvA/VPrfvkxjk2sri9uBNPeodHtN1o+KyDKlvAjPIaxslsb3xfmLHSwRpwgznqmAXOoicm4dWLJGz4yphkvMmUA+w/Hw5uED9mj5/q7FTzj68WTg0ZgmR0IFomyIUxlVmT59tDhKVzjbaurddeF2kkCsN5JoH+CFHk5IazCBCtAKwlG+4H7UmVFMS1rQDk1Vq5vt2IpzsGCZtfq3HpOBjHxPV7wz+Btf0+p//+R3JvutemrL7zmYfbIpwrG2eEaGrzVpXtjTSTQORG0gdikEDsCa6VghQYoTPeorOxhQmFJL+Zjj41mxGv3bWYlqoj+bfYSK11BWYpm3OnokDaqFMmEjWrwO3w9VdEvP0NG2kB1HLn2LuyJHjnS90P+e1J8wQkS7nnb9pcR6qnqJovSxSbBiDT+W4l0ky5vMuFpEKlCG40x90E3rCiThaDSkdG7FTSCtIV2mGy7RLf7+3kF+yXvohi6hyeYfXGxDG4aKs+FcXW9mDKHmf2bMdzl2U465bMikGPEf6T3eqDwGfHLaOutqqp0wm3c+eMuNiNDEGbSV6dF9HtvkWAyd/huOrphd+WfgGdHRR3SV5BCbXxiBM0DVM5ETSbgjjOEIjzuzWCxebXhYdCHQ/HgU614HngW5BU14QHiiDe6zvxtb5XXywfojaJ/qyovg0C5hicJQSDhrCVi/q/xAWCfk+3qJ9Z4AF3pMSof8TDZHgvmP5mH0QSwbR3efiUDKwiS6Rv377V0iz2OdEedCXS1gMXMn2yExxiRHOeGV7bbW5zfxcWeqYQdN9XYqRdm2bsT9m58Z9LEeVHfknlBA/G49TlXtvlsHLsjvB1HcxTQf7rQGLbCKO6YnHNqIuymIlPiYHq69IfDyNBVMhmpyprUvZPEPa9SyZfBk9+0L1/l+UFHxkUDSbtqySpJFtPM1TKR+quupDnkZ8WZP+SePmLMiJOlJ4OZ8a7ajrbMuQZQQGDe2nMIJLfwMTC1f/6bflq6xeLvRrXUgi0prCz4pQcSeKY5noD/2xhCYvGcuYyGSlSn5ll9OVya20awAopcP7DaFZfIRkuoPRyzwzl9auV2z7uKYnxaJNcodq28mG3Is+YMuxAccNs/eX71COg6BJqUF75jkzCAlrcuySBMQISCQ+Aj+Yo/e7+QT0UuaFZF7X+/H6vXkl0YYCayYfmBiX9+B/m9cqZGUZK6Zcr46xgIqoGwDtXxJm9V2uLOrMN1fk8N/ze4HlZ3nIEfD3hUu+AljCDTxlKIY0ej4Jlok3fYrSaG8KlExh4mcr+l5SkyxBCVsb38kmOjynBeZPQzYN29mDtYNV1C/SIYSlOv38rvAZindURIIZ9AcatQJIgo7rk5/TTtDNRx6GChhdnQq+1N+J6xmGgC8p7wMnd/M4zKXag9eyMzUYuGq7X28QKzzrT5UdqmTnJI9prpAp1/4AG4EBWMB5QDxSpTUMF1YehnBLcCzipP/9A8wdte2K89MfVaF06hH9OtWHrsmArcA+mqtFaKU8oeazmwgg05Nd6fcUG+HBVki2a4fQyjHnK37XwlrAwIyAEaL7ZJfw2rIvt973Eywx9krP0p0ZWwc2vO1IBfWQuN7FgFy9zNwPNY2WRnDfr7f+9g0CrSL6OpAsC1phSsZvXlANCThp+RSWF7peolmGyK2JEMK2SKB09u4DGJL+0s/qQid2rWjB2YCSzbOBoQ6i6JaxQ55Xewo5yKzxUTmjBm0FCEpFAcEWZCYH2VIAd6fdjyj7koxoX2li2KQsRlc2UBheuDKP1dZfPznW+8pwLH3WuUJVUlriXFF0rRrj7bwbHVIIQJAL04dMM1NOSPx9S2hw/qMyt5gFAa4/JOeXhfOGLjmDOb4XyInO1plgt46i3noRVSzU3fLnnBLRTb0Zy4BG5vj0Sw8wGUXu9Ka7BfnHCkbPdfDKTbfMAOm7ASqRkeDS38jUcMpSxWciqkDxZlUXwIJ9euXKqbyBpGTMQjJYgH3t5g2L/dCrut3zJXNYSxCrHquHLnzKkdHn5i7qTu1Jlk4kGQ5IkbrmS4QWyJ/sD+aao2qtvWQFzH/OhMXBAtkrq7weHXmLMOsL8Fq1XhgVfXhMhGVgoS5kSlnZTc+c41oeLF3gxJ1HkxmmkvMcvbmMmJRnxoYjtZvgGK5ZhiBtWGjvs/eiVW0KnWdbUE8yDBtk8f53i6YbTdSoAhjvMS9T5xPnRs0cFHCrx5j7zubrq+bTEUTCil+8V4CYzaHpppHFSBP88U2HF7XtzjTGs7pUav8YFSOcM7eQQlCc7CgEsagfvejJLL0o9vRFXgLf19m3CoiJhk/MHEOAd2gCBc7L5IU+NXFg4cFtLPsR366E5ujWuTQAg+bDxbiayuuo9gw1Ed/ZGGa7orTKtTKuronyDnkCEgKvnAgGS2e9EfWEby2G0+JLK1oK5vh3usYeyINe9RPc13RAw+HJa55+0iFfsaxNongeUX5pa7MOJPoVkuE9PJoUj11rwMVlKnF5Iq6XYSPwdlqM7ep8ByXOLB7rkxg+MWMyCD+uaDg8+IT+1NVd2RNVwgIUX7qBEPBHgu00goaj4Xnxn+nO7Xfudozt1alk3ddZYUuE3OZATInOxXRApUn2x6Awl4mN8oxVVz/oNefAyUa/uBv5ufdJ2ywGRbjHOelDt/kisJ+Qckf7yLad/x+BobVu9c+4ann4UkC6zF76q+C08ixmrSRVJjQ9TZfzYS3xjt0+0sWoFamJAiNFlHwupoo+Z7BWhdaRyL5LuLHg3qxS1rmaVG3wRVK3WXnwq84hxtWRFfjpdCuaApuhzDmChvF/gN8km3gc3atc14Vaq5K0Y+pdRjeIVM1ZsaoLSWli91SI5+CobNGGo2D4vbaay4RtoDtOtK6zmVRa3R6BRK0PfpXLiKJIcG0NNzl9Q1hPeYvEe457sJ0v1UCP2AN10qVL34oY2P6id/yRJhvqyxSkwS+gQxYjT4BUqP4DCviCKIPsq/cq16dJD6M18OdnU5ciBWDlbTXYBxsPLjbFEIPNGeMMshC8i8nD13oKWcsj+RzQgVl5G2lvM6w0L1IyRXbqeDr5iH1tTV6VzIzjyl1k2rJR83h6qHxSuykj0zKEftkhrluPyEx6ndBgK8VE5MqhtFAwphha+4iVABkHGfQVZf0f7LMn7VlTU4jvUPTU+pFLIPlIsdCMCCfRNCF3/yIO9DnvY/cx5i8bzZDPPWK5vM3QP+nZJ7CI9Po9gHIp4uWuW5AkDDz9ybVr1/y/kuU8p9/++7aeKE6oeiic1Wsypu1ZKnVkvlLsgMpQl68YogfuRXbEjqsy7p4TE2fIOa3xJ9OBu1XhZ+bK48zfCi1/yqR339tO0Y+aHlkyBxmi9YQzs+zFOVipcIT0emB9vxBCq+iliOUnKDnKFsRUAmExQi4wRiXA4Pi8eOqBBf1IxJLs4nIaoV2qzhjbrlTL8pZ2UZzXTrgxbG/a+J78AsLuR/LCDqcEC6i3vchwiTVZceYT+a9xLSzqVP5BoJ38lwWIHYUM+jcyH2k1cForl/k/JLSFKSkXI1ADhsqYYHmnr3zk1dzs2EhBie7vKQ1nmJbuYWW6OeiBjtvezUBK4ogPpGdbB4TIcHV8K+lVBX9V51q+ieco4WaS4qG6/msG82c3EWq0UztOaFA0U+aG3IwmADE+Q2oYRV9qLq1LQJ7AqBIjsruvSBD+8cq4Au4jZOkahmIieDUIwUoK0DPYZfoY8XbWqCM/DZSY7zlx273SqzSqlgVWR34gtubaO9yPMJ6XPgoBV+i8ox4WOcnkawv2h7tRDjKEphOFJ739X74JxIIOvPYEX29Es9vHSIbL2mepC6fFPB2gcNUSIKpC8tqsFdqhix3OubQ9HV1Kn6naF2pSpfmSmpTEnzV5p/MzOeIqsj7Cv3jjq+dZ+kxl9asIIF8iLc5PC0cvNpU0JF2gtVQYLS5V6ggRaIKAUd+gm0l65yNI2vOpk12ttCry6KmFgAmzMDk1uS/JRubdKNtb/r8t7yJ6EDyiAIcU1y1P3nJp3iLMOOCgvUA84KzCSImDfw4gplz06Bn93ULrI1n9ZB9yjCyRmx5S7LhO3IGbw5r5pav3Rtaz6kp4f58OepwbIq5oj2Ph5N2A+/I3+aDltecjHrQ7oU54hisS5QjPukVQwL01bAGW9VHja0jl4wjaJW87Ia+x0qEMrl9CtVLHrBXIXXczkXvZeoeIZV88sb0a1e0rUlOlTz4GuWeTsr8bRcs6l57YrxypVEReEe+0Iq25CknL1L5kmKTyZWq1H+OZeEMR5EfwcpQfF/by3ek9YhxtbyzjaHZExRWppoDvcCSt3ThJvDSdJO0ZYysm3iLImjy8VrXovndsCX3apC9Qo9uWBoUQGPEtCzSjNMpwh7LEjFXlhJ0pNKgyjhieb5vxY0KwKX3mQvURwZhVaGNimn69Ov6il/dvbyjTuGuPtwyqO4VOmng8cfEbwlbWljATrVfVeGq5BxQOOYVqHJBiAKgU37GNFGYHbvYOnTcbzAbbkEpGKGyOQmHESA6s1+5VwsxL1M05Bk2dhV27LJXdbcTUIRZZQtZcVPhCtQ8G4V7bMqmEyo/IykUXd4Y36pxJbxnLTgBRgjUgYjcbTeMFwOpTkOp9do9IslO/3rdW2GiiBcROeHZFMJHwz4+HT+2gmEMl435rXhcXId4m891pVfjV6byQI0TgHqKpoR9i71DLoI+QEi3y/SBq/zNgaJoSEnzbwFtyqMlw4vfwuGenYMFpnT8+xrzhc/yR3bucjUwSKUUvNL3ECBQDp3KLIhgzIWEglYfJgH0RbY2c6g4Hiy9DjCAvROshuNp3xuA+r3Xp5F5Hlo0Zy9ROP3OO21W7Rc2+TW2Gq/XTQZ59VInkb6WFo9fbBUW+UGCusTogL94H6VlR/U+DYezETAJwtNNORY5dByBdI5iFEt/Cqmaqe2BDpybNLYtBcDm3XXReYzyoZ3Yj9Vyy4gutZKm3EwmtfPL8qa1Rx+4Bomj5NV9cbObEuILd/7k5vINyURRaicklCvsn9j4vrQ8/sJ6qhHB34zNh5dN/pZXmTmIi3ii974Dt066cRT8pt1mPn3VnaxQ97hyI+jo7YVIhHg2L+Ux4OGnpsv7gK9jkapbflfRJdYLPatAIq8b8dpLJR4pYs9fsLYIJEHNJXkFSFt4sRgD', 'page_age': '1 month ago'}, {'type': 'web_search_result', 'title': 'Otter | Description, Habitat, Species, Distribution, Food, & Facts | Britannica', 'url': 'https://www.britannica.com/animal/otter', 'encrypted_content': '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', 'page_age': '3 weeks ago'}, {'type': 'web_search_result', 'title': 'Otters, facts and information | National Geographic', 'url': 'https://www.nationalgeographic.com/animals/mammals/facts/otters-1', 'encrypted_content': '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', 'page_age': 'May 4, 2021'}, {'type': 'web_search_result', 'title': 'Otter Fact Sheet | Blog | Nature | PBS', 'url': 'https://www.pbs.org/wnet/nature/blog/otter-fact-sheet/', 'encrypted_content': 'Ev8QCioIDxgCIiQ4ODk4YTFkYy0yMTNkLTRhNmYtOTljYi03ZTBlNTUzZDc0NWISDPmFppA7LiuhUBavehoMAHRiVN+n+RonnTm6IjAAYGLASmjS4wj+FeDI7eyOlA0wcrR+1ezb9RXa3IFgVrURJ1FVWS1Nqp6mXYBeEZQqghDjDzACZ6YeWFnQcJ2HjSVLPYrUaSESwjNUR7bBbsbWxqHoB8uoqOt5lzg1ZvTWuJZxnv2Og6zQgJho7ETkNKjl2FvqqYX7Ifx0Qo/3SGEmbQU9LbEhNh9MrLrkM7x5WIekcv5L/ueEUd2tF9tv/0hTVHAQwWMyg2GoA0C6+23DdLTFyad2yHmxsCHRMux3JW2GD/IwJjVLhCmrHWab6uLi90tDNyf94WpyzMWZwDn2/0Gbk7epRSqMp81gGk49vWyvnfwLRc/779gIM1Zo6NaeDpUq7T94lSRKAQf1GBTDN3CAQnF7h7RdrutpUvrjHm5iW3OedeBol+WsJsqazrkNgMgMwJ1TWyMPhsmhDIm9a6+TQymQuy8ZIC7nzszO8835UHR1je6kNeyXFAPxSdmcWZdNyhg4yadCp372zkC4HCOdnbo8oYKIflhzcFlFc8r7YqF/SsgS8EA4CRYSn4ZSA0lOFJByKodDOxFDW2ppDmuv+K0GJfNuA1ytTPuVVBHR7WL/hB+M1wsvB6LyuMODaXrhOil+kk0iUuFM6wnn+yKEwBCNnb+cQDiQdR9/ZVVz69GResMG1fnkTHli+IAd9qnoGbpjgbyiPbVU5WisHwY9luW7HoQWK2yQ7gh9VWHzqAdqLZtnWV/SujEQk/Q7U5IecwlYbAWnQf3hN6ff1MOYLofv8xcL8ubtS1yJXCI0o2E8x3oxs8LASpOqKhQvbQ/sYeBlxgcrqEbo0SPgn7AfboP4C6buXrFLT8W3elnVczct3PUK+mPoQtZMMnoa84yNYHknJ5ncR8pMMyfVzYaGFD6Y3k0OGkPZ8QYCesGJQP1QhhayQcUroAihia5IPiGn+owFmEdBI4njRn5WzM3YmEsEVeD6Dh0flg2mAiFjazewwHaDZIx09G7aGT9xfeh//4Ko7vpJpsAkn9Fd1DgVE4RmEDO0975ZkXXYUX6frW0CG8//ou8JomMHH0vHVOg319QBPc2wedet1OEIfwGizJay/eHC8Vy9p3VkcvIMeIkdCxWexAphuqiOounVhAiXAqiRqWm7QMoZpe/ryrq+j+2nj6SKhDOXnsW1OcWvz+jTTu60uEhDZiRitK27wBTLk/GXRJhqq5wAJQSKgF86XyPZ3TavJLKifyYpP6T4QtoY+UHk4km3T6ZkowwPPR6T0pVuJXA3OB9QNh5sC0RI9JIXIebOT3+X6hzujy9HerDeKAZPKbScz6mn4g9W8366Dw5qCfUclCzjTeARa3znawNKDgtzImNRTjaugF/oq5fE33GDx1H80coq4gdZ0frmQSnjZ0MUoHzhhnYlAnSp1ACdhdt+Lwqx3jGU4Sz0vs/15ILbLu4YdRCZ6Kkb20QVPopx5xgtPFbHY1M8/EEJAWFqOPWeLTHQxm0d0ED2AbWY/e4CyCM23mDbjWSfrEi9sB+3BXHrNJgziQSjmpL6T5hYwXgOsWH6lmmg6287CDIvC759bVZQ4ciqtrr2I7Pcd4vjjNQt9Vwkl96xWN2yNLIywRfJxJQFAN7dQTPsq2wqq+sw1WJz9Ntavm2JlnPYF0uYCR9fFQ4X9HRAPNuOG+Z0J5zuOGVVE5q5FrTGnPK7ZrfPAQxM4PYFyUnp0hjVLT3yy26abbRoPQWLOw55RVhGorgjcvbVY6mNJPh8ZL1DUcD9F12sG3y5KbondH63Uve+E0aU3HRi/Bm6GyyhN7JxogEVdGSjqFAhaZYYcqIHTINVI/IKXcbAqKmAs9g+xiA+BzWnlCrdBkuX8XZbjdt4o/dz3JHxK+tSWThHhckD0druXqqDhY5j79pzdAGj8VKD70GRUs8KW3Osua/ww/8krzVof8tMIyVZlyMtw5QIV8vjdMYqEal7cqusr8jYDMd3n7t2cl+jpBsDfiTMYo/Uw7KD2r1u4huc9KXWrUCLnx5ZkT/V1OcEkb4QkCWsdSY1qXHGnirzKS5i8XeuWJuuuEHT+X7SJZKTVpkwcnxdFLRxrxolUaqwVXGIoD9PD8T3p3gVgLb6YbuKeNNuGAesWQXX9gLlXO5me3Na3R3+1Dt+G22BsllJK1pHwPeUyo7wlm+BRTtti54TxhQgdak7ZneTfotjvAh0e49GBeR7iBaj2VcX6QG5DzxN054qsugsu853HgJJVx3C4EvF9xzbrQDIB3l/liIompbplLuqBH+yLOcamU/c87QHieJRPXW4P9THeqcDh8KWw8zBAMxRqxmf1L5UXJF07RIqs+xpmlL9XNYbVdyHeA6z1prMvby7aK8JKV2KAsNpRIsKHAHvLyoBCc2NLuVCiAOUP08AQtwmm1Aj9cuGHC71RtEPXTuiB8PZRg7mIF7GIJAP1GNUNRNkmQ6eEGObqb+rWyWgub7nGJPY9MRMBX38yV7RlJUGUjzNTdLZPT6TtFXNutkgEZbFga+xWsuGUAJmjKjMf2/Yh3oqUGHVVitZDzol7S+kfPo4lb+j1NcggLPuAVlughJpLtURhM6AgJlppQg1TXPYwe9XIZzYgwVFyfpV9QaZ5QO4GCIExWVEan/1lCKGiliVSqpwTR62ZgWI4WdaYw0mmx/6togBA/xFC7srvBN6mExLcjwmqLRSKu5q5wIhr52xrjtExHGGQnCSSPIk/5viLRMADjFpr++Yr0zVCJgPg/y3Tl9pLVNVHvUTKXM1uzPyokj5EN/F5J/nmttQxbu29DXCba9Jdjn4oEThSeSJ4qgnYbLeUUh+FBkCGAM=', 'page_age': 'February 11, 2026'}, {'type': 'web_search_result', 'title': 'River Otter | National Geographic Kids', 'url': 'https://kids.nationalgeographic.com/animals/mammals/facts/river-otter', 'encrypted_content': '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', 'page_age': 'October 6, 2023'}, {'type': 'web_search_result', 'title': 'Seven Quick Facts About River Otters - National Forest Foundation', 'url': 'https://www.nationalforests.org/article/seven-quick-facts-about-river-otters/', 'encrypted_content': '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', 'page_age': 'November 18, 2014'}, {'type': 'web_search_result', 'title': 'Otter Fun Facts! - WonderLab Museum', 'url': 'https://wonderlab.org/otter-fun-facts/', 'encrypted_content': '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', 'page_age': 'February 20, 2024'}, {'type': 'web_search_result', 'title': '18 Otter Facts From the Playful to Slightly Bizarre - TRVST', 'url': 'https://www.trvst.world/biodiversity/otter-facts/', 'encrypted_content': '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', 'page_age': 'December 7, 2024'}, {'type': 'web_search_result', 'title': 'Otters • Chesapeake Bay Foundation', 'url': 'https://www.cbf.org/nature/otters/', 'encrypted_content': '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', 'page_age': 'December 3, 2025'}], 'caller': {'type': 'direct'}}], 'citation': {'type': 'web_search_result_location', 'cited_text': 'Otters and their mustelid relatives were once hunted extensively for their fur, many to the point of near extinction. Despite regulations designed to ...', 'url': 'https://www.nationalgeographic.com/animals/mammals/facts/otters-1', 'title': 'Otters, facts and information | National Geographic', 'encrypted_index': 'EpMBCioIDxgCIiQ4ODk4YTFkYy0yMTNkLTRhNmYtOTljYi03ZTBlNTUzZDc0NWISDMIIEh2AaP0UlVZkzxoMBitXGNpXydPJqYXPIjAjhZaLLwG4+yA8RTTcdSjDcSYHs1pSCDehQsV7gd4sPqBg18Ly1CIzqXkBNHEWlbkqFz/hb6VQ/DCCzxa1ZJwbufPU45VCLY8zGAQ='}}))], usage=Usage(completion_tokens=627, prompt_tokens=17556, total_tokens=18183, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=627, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=17556, image_tokens=None, video_tokens=None, cache_creation_tokens=0), server_tool_use=ServerToolUse(web_search_requests=1, tool_search_requests=None), cache_creation_input_tokens=0, cache_read_input_tokens=0))"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gpt-5.4-2026-03-05', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='Using the `simple_add` tool, 5 + 3 = **8**.\\n\\nExplanation: addition combines the two numbers. Starting from 5 and adding 3 more gives 8.', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'refusal': None}, annotations=[]), provider_specific_fields={})], usage=Usage(completion_tokens=42, prompt_tokens=199, total_tokens=241, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=0, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None)), service_tier='default')"
]
},
"metadata": {},
@@ -6217,349 +6792,253 @@
}
],
"source": [
- "for m in ms[:-1]:\n",
+ "for m in ms[1:]:\n",
" display(Markdown(f'**{m}:**'))\n",
- " chat = Chat(m)\n",
- " res = chat(\"Search the web and tell me very briefly about otters\", search='l', stream=True)\n",
- " for o in res:\n",
- " if isinstance(o, ModelResponse): sleep(0.01); display(o)\n",
- " else: pass"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "7dbab35b",
- "metadata": {},
- "outputs": [],
- "source": [
- "m = 'claude-sonnet-4-6'"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "18f95e5e",
- "metadata": {},
- "outputs": [],
- "source": [
- "def mk_pause_web_search():\n",
- " srv_tc = mk_tc(\"web_search\", query=\"Solveit Answer.AI\", tcid=random_tool_id().replace('toolu_', 'srvtoolu_'))\n",
- " pause_msg = mk_tc_req(\"Let me search for that information:\", [srv_tc])\n",
- " return ModelResponse(choices=[Choices.model_construct(finish_reason=\"pause_turn\", index=0, message=pause_msg)])"
+ " chat = Chat(m, tools=[simple_add])\n",
+ " res = chat(\"What's 5 + 3? Use the `simple_add` tool. Explain.\")\n",
+ " display(res)"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "f7c25d76",
+ "id": "62fe375b",
"metadata": {},
"outputs": [
{
"data": {
"text/markdown": [
- "Let me search for that information:\n",
+ "\n",
"\n",
- "🔧 web_search({\"query\": \"Solveit Answer.AI\"})\n",
+ "**gemini/gemini-3-flash-preview:**\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(**gemini/gemini-3-flash-preview:**)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "\n",
+ "\n",
+ "🔧 simple_add({\"b\": 3, \"a\": 5})\n",
"\n",
"\n",
"\n",
"\n",
"- id: `chatcmpl-xxx`\n",
- "- model: `None`\n",
- "- finish_reason: `pause_turn`\n",
+ "- model: `gemini-3-flash-preview`\n",
+ "- finish_reason: `tool_calls`\n",
+ "- usage: `Usage(completion_tokens=84, prompt_tokens=85, total_tokens=169, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=66, rejected_prediction_tokens=None, text_tokens=18, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=85, image_tokens=None, video_tokens=None), cache_read_input_tokens=None)`\n",
"\n",
" "
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model=None, object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='pause_turn', index=0, message=Message(content='Let me search for that information:', role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, function=Function(arguments='{\"query\": \"Solveit Answer.AI\"}', name='web_search'), id='srvtoolu_N0pThnXuYgjRcJ4HQbNLJyw8c', type='function')], function_call=None, provider_specific_fields=None), logprobs=None, provider_specific_fields=None)])"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-flash-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=None, role='assistant', tool_calls=[{'index': 0, 'provider_specific_fields': {'thought_signature': 'Ev0BCvoBAQw51sfwHaoJJ2zAYvG3OjGxnzXwweSDUEiPQ8Fctb+Yq9H/nGaNfcP9TiUmldLfuoOpVqJR6XarYK1JQRU/pctffHmGtm3wUl8Xmvfg1LnyftTobksKcjUPFRlcZwJGdHI3lLGdh67Gh3WgtgY3OoXWsCMI37TaI+8qEtLl+6XLJbahLFlJXnzwCixYRxA2j97jNG6YWQpdYk5ji1Zbd96VTSiwhiC84SQorDWy1o85hLbxVHj6VMG1Aku+SO/sl4wyMZCLCX2ZPZuRYfgR49qaCJk1TI9jA9yQL4WB0oSdZ1Rw2ATPfG2um9AWLnCmRiUE97DIUFaotQ=='}, 'function': {'arguments': '{\"b\": 3, \"a\": 5}', 'name': 'simple_add'}, 'id': 'call_09863ecacead4753a32ec8d54d9f__thought__Ev0BCvoBAQw51sfwHaoJJ2zAYvG3OjGxnzXwweSDUEiPQ8Fctb+Yq9H/nGaNfcP9TiUmldLfuoOpVqJR6XarYK1JQRU/pctffHmGtm3wUl8Xmvfg1LnyftTobksKcjUPFRlcZwJGdHI3lLGdh67Gh3WgtgY3OoXWsCMI37TaI+8qEtLl+6XLJbahLFlJXnzwCixYRxA2j97jNG6YWQpdYk5ji1Zbd96VTSiwhiC84SQorDWy1o85hLbxVHj6VMG1Aku+SO/sl4wyMZCLCX2ZPZuRYfgR49qaCJk1TI9jA9yQL4WB0oSdZ1Rw2ATPfG2um9AWLnCmRiUE97DIUFaotQ==', 'type': 'function'}], function_call=None, images=[], reasoning_content=\"**Calculating the Sum: A Quick Task**\\n\\nAlright, a straightforward query – the user wants the sum of 5 and 3. This is a perfect job for the `simple_add` tool. It's designed specifically for this kind of elementary arithmetic. I need to feed it two arguments: `a` and `b`. `a` will be 5, and `b` will be 3. Simple enough, no need for anything more complex here. Let's get this done.\\n\\n\\n\", thinking_blocks=[{'type': 'thinking', 'thinking': \"**Calculating the Sum: A Quick Task**\\n\\nAlright, a straightforward query – the user wants the sum of 5 and 3. This is a perfect job for the `simple_add` tool. It's designed specifically for this kind of elementary arithmetic. I need to feed it two arguments: `a` and `b`. `a` will be 5, and `b` will be 3. Simple enough, no need for anything more complex here. Let's get this done.\\n\\n\\n\"}], provider_specific_fields={'thought_signatures': ['Ev0BCvoBAQw51sfwHaoJJ2zAYvG3OjGxnzXwweSDUEiPQ8Fctb+Yq9H/nGaNfcP9TiUmldLfuoOpVqJR6XarYK1JQRU/pctffHmGtm3wUl8Xmvfg1LnyftTobksKcjUPFRlcZwJGdHI3lLGdh67Gh3WgtgY3OoXWsCMI37TaI+8qEtLl+6XLJbahLFlJXnzwCixYRxA2j97jNG6YWQpdYk5ji1Zbd96VTSiwhiC84SQorDWy1o85hLbxVHj6VMG1Aku+SO/sl4wyMZCLCX2ZPZuRYfgR49qaCJk1TI9jA9yQL4WB0oSdZ1Rw2ATPfG2um9AWLnCmRiUE97DIUFaotQ==']}))], usage=Usage(completion_tokens=84, prompt_tokens=85, total_tokens=169, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=66, rejected_prediction_tokens=None, text_tokens=18, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=85, image_tokens=None, video_tokens=None), cache_read_input_tokens=None), vertex_ai_grounding_metadata=[], vertex_ai_url_context_metadata=[], vertex_ai_safety_results=[], vertex_ai_citation_metadata=[])"
]
},
- "execution_count": null,
"metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "mk_pause_web_search()"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "b3a9f608",
- "metadata": {},
- "source": [
- "We mock completion to return `pause_turn` in the first 2 api calls:"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "95067f82",
- "metadata": {},
- "outputs": [
+ "output_type": "display_data"
+ },
{
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Mock Call 1\n",
- "Mock Call 2\n",
- "Mock Call 3\n",
- "Total calls: 3\n"
- ]
+ "data": {
+ "text/plain": [
+ "{'tool_call_id': 'call_09863ecacead4753a32ec8d54d9f__thought__Ev0BCvoBAQw51sfwHaoJJ2zAYvG3OjGxnzXwweSDUEiPQ8Fctb+Yq9H/',\n",
+ " 'role': 'tool',\n",
+ " 'name': 'simple_add',\n",
+ " 'content': '8'}"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
},
{
"data": {
"text/markdown": [
- " Based on the search results, I found information about Solveit (solve.it.com), which appears to be the most prominent result. Here's a paragraph about it:\n",
+ "5 + 3 is 8.\n",
"\n",
- "Solveit is both a course and platform designed to help people solve problems using fast short iterations, covering areas like coding, writing, sysadmin, and research. The \"solveit method\" is a modern approach to building software, writing, solving problems, and learning, inspired by George Pólya's \"How to Solve It\" and developed by Jeremy Howard and team at Answer.AI. The method is founded on building in small steps with quick iterations and immediate feedback, and for coding specifically, involves writing 1-2 lines of code at a time and immediately showing the results. The platform provides users with personal instances—full virtual private servers where they can install software, store files, and host applications. It serves as an antidote to AI fatigue, helping users avoid being overwhelmed by AI-generated code they don't understand, and has been tested with over 1000 preview users.\n",
+ "\n",
"\n",
- "🔧 web_search({\"query\": \"Solveit\"})\n",
+ "- id: `chatcmpl-xxx`\n",
+ "- model: `gemini-3-flash-preview`\n",
+ "- finish_reason: `stop`\n",
+ "- usage: `Usage(completion_tokens=8, prompt_tokens=289, total_tokens=297, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=None, rejected_prediction_tokens=None, text_tokens=8, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=289, image_tokens=None, video_tokens=None), cache_read_input_tokens=None)`\n",
+ "\n",
+ " "
+ ],
+ "text/plain": [
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-flash-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='5 + 3 is 8.', role='assistant', tool_calls=None, function_call=None, images=[], thinking_blocks=[], provider_specific_fields=None))], usage=Usage(completion_tokens=8, prompt_tokens=289, total_tokens=297, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=None, rejected_prediction_tokens=None, text_tokens=8, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=289, image_tokens=None, video_tokens=None), cache_read_input_tokens=None), vertex_ai_grounding_metadata=[], vertex_ai_url_context_metadata=[], vertex_ai_safety_results=[], vertex_ai_citation_metadata=[])"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "\n",
+ "\n",
+ "**claude-sonnet-4-6:**\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(**claude-sonnet-4-6:**)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "Sure! Let me calculate that for you!\n",
+ "\n",
+ "🔧 simple_add({\"a\": 5, \"b\": 3})\n",
"\n",
"\n",
"\n",
"\n",
"- id: `chatcmpl-xxx`\n",
- "- model: `claude-sonnet-4-5-20250929`\n",
+ "- model: `claude-sonnet-4-6`\n",
+ "- finish_reason: `tool_calls`\n",
+ "- usage: `Usage(completion_tokens=98, prompt_tokens=610, total_tokens=708, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=3, rejected_prediction_tokens=None, text_tokens=95, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
+ "\n",
+ " "
+ ],
+ "text/plain": [
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content='Sure! Let me calculate that for you!', role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=2, caller={'type': 'direct'}, function=Function(arguments='{\"a\": 5, \"b\": 3}', name='simple_add'), id='toolu_01PRYuuvarMbffFgYND4aeGu', type='function')], function_call=None, reasoning_content='Simple addition.', thinking_blocks=[{'type': 'thinking', 'thinking': 'Simple addition.', 'signature': 'Es0BClsIDRgCKkBlwmZMRjyibQQ8AWRB5B+Lv74nqa3VUhZc6QFVjEJCrtYPiJ7f2pG54Q9kV4YS4IR6Maq4kXu/ObtRgkBswbg7MhFjbGF1ZGUtc29ubmV0LTQtNjgAEgzBvG1fa54AO3yik5MaDKuOaXK3MUk3rGHsMiIwTT/4Ol2p0Di3DTpVxOjoMt1knuAH6+2jsMr6LFHSf8vs6nxTooNwuSGfmWHGpobsKiAzrQgo8XY8LoEKlI4WGIDJ3qdVaQWOSJSwk7fXA2003RgB'}], provider_specific_fields={'citations': None, 'thinking_blocks': [{'type': 'thinking', 'thinking': 'Simple addition.', 'signature': 'Es0BClsIDRgCKkBlwmZMRjyibQQ8AWRB5B+Lv74nqa3VUhZc6QFVjEJCrtYPiJ7f2pG54Q9kV4YS4IR6Maq4kXu/ObtRgkBswbg7MhFjbGF1ZGUtc29ubmV0LTQtNjgAEgzBvG1fa54AO3yik5MaDKuOaXK3MUk3rGHsMiIwTT/4Ol2p0Di3DTpVxOjoMt1knuAH6+2jsMr6LFHSf8vs6nxTooNwuSGfmWHGpobsKiAzrQgo8XY8LoEKlI4WGIDJ3qdVaQWOSJSwk7fXA2003RgB'}]}))], usage=Usage(completion_tokens=98, prompt_tokens=610, total_tokens=708, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=3, rejected_prediction_tokens=None, text_tokens=95, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/plain": [
+ "{'tool_call_id': 'toolu_01PRYuuvarMbffFgYND4aeGu',\n",
+ " 'role': 'tool',\n",
+ " 'name': 'simple_add',\n",
+ " 'content': '8'}"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "5 + 3 = **8**! Let me know if you have any other calculations! 😊\n",
+ "\n",
+ "\n",
+ "\n",
+ "- id: `chatcmpl-xxx`\n",
+ "- model: `claude-sonnet-4-6`\n",
"- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=386, prompt_tokens=12333, total_tokens=12719, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=386, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=12333, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), server_tool_use=ServerToolUse(web_search_requests=1, tool_search_requests=None), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None)`\n",
+ "- usage: `Usage(completion_tokens=27, prompt_tokens=721, total_tokens=748, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=27, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
"\n",
" "
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-5-20250929', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content=' Based on the search results, I found information about Solveit (solve.it.com), which appears to be the most prominent result. Here\\'s a paragraph about it:\\n\\nSolveit is both a course and platform designed to help people solve problems using fast short iterations, covering areas like coding, writing, sysadmin, and research. The \"solveit method\" is a modern approach to building software, writing, solving problems, and learning, inspired by George Pólya\\'s \"How to Solve It\" and developed by Jeremy Howard and team at Answer.AI. The method is founded on building in small steps with quick iterations and immediate feedback, and for coding specifically, involves writing 1-2 lines of code at a time and immediately showing the results. The platform provides users with personal instances—full virtual private servers where they can install software, store files, and host applications. It serves as an antidote to AI fatigue, helping users avoid being overwhelmed by AI-generated code they don\\'t understand, and has been tested with over 1000 preview users.', role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, function=Function(arguments='{\"query\": \"Solveit\"}', name='web_search'), id='srvtoolu_01VCLaffd3zxyZdvqNV1xkQ2', type='function')], function_call=None, provider_specific_fields={'citations': [[{'type': 'web_search_result_location', 'cited_text': 'You can now sign up for Solveit, which a course in how to solve problems (including coding, writing, sysadmin, and research) using fast short iteratio...', 'url': 'https://www.answer.ai/posts/2025-10-01-solveit-full.html', 'title': 'Launching Solveit, the antidote to AI fatigue – Answer.AI', 'encrypted_index': 'Eo8BCioIDxgCIiQ4ODk4YTFkYy0yMTNkLTRhNmYtOTljYi03ZTBlNTUzZDc0NWISDCC8srdB1wUvAaXEPRoMMKcmTvV4OJxqSmdSIjAJ0zd6pe2BkLp2sdeDWX8/x0JbDgCEa4pObOskY9bPbPM/vik9iTtCdibVIC3bxKkqE0WdsD5kpeqsl95k+iIfsvEMgfsYBA==', 'supported_text': 'Solveit is both a course and platform designed to help people solve problems using fast short iterations, covering areas like coding, writing, sysadmin, and research.'}], [{'type': 'web_search_result_location', 'cited_text': 'I want to share with you something that, for me at least, has turned out to be an effective antidote: the \"solveit method\", a modern approach to build...', 'url': 'https://solve.it.com/', 'title': 'Solve It With Code', 'encrypted_index': 'EpABCioIDxgCIiQ4ODk4YTFkYy0yMTNkLTRhNmYtOTljYi03ZTBlNTUzZDc0NWISDBmdiaXap6wP7kJwdRoMDYQz1iPEIktn2YOtIjAUanryfHdCztyK33RnWxrT28uAAEWjIJErZTAQQlFpUtmHN/HahQEeKIeNA/As8x0qFLaF22v3fK5MNCiFqyPs2sdT0oS9GAQ=', 'supported_text': 'The \"solveit method\" is a modern approach to building software, writing, solving problems, and learning, inspired by George Pólya\\'s \"How to Solve It\"'}], [{'type': 'web_search_result_location', 'cited_text': 'The Solveit method is founded in building in small steps, with quick iterations, and immediate feedback. ', 'url': 'https://solve.it.com/', 'title': 'Solve It With Code', 'encrypted_index': 'EpABCioIDxgCIiQ4ODk4YTFkYy0yMTNkLTRhNmYtOTljYi03ZTBlNTUzZDc0NWISDOAwBEDkmABorHfSghoMNbC4vObuxsQdOGZsIjAc9p50E3UkZKbpGLYQH7UOBA06WIFbdYKLvt4Qu6p5ZaaAHqaQHZekTfGsxvJYwlkqFI/Ng3F/9fNpBBs6ciOBsDm0eLWMGAQ=', 'supported_text': 'The method is founded on building in small steps with quick iterations and immediate feedback, and for coding specifically, involves writing 1-2 lines of code at a time and immediately showing the results.'}, {'type': 'web_search_result_location', 'cited_text': 'For instance, for coding, the approach is based on the system we have developed over the last decade with nbdev. We write 1-2 lines of code at a time,...', 'url': 'https://solve.it.com/', 'title': 'Solve It With Code', 'encrypted_index': 'EpMBCioIDxgCIiQ4ODk4YTFkYy0yMTNkLTRhNmYtOTljYi03ZTBlNTUzZDc0NWISDE3ReXcCBPNSTlIIVhoMK70aJSxkw26doD4zIjCq6ne7VA6qrnt+8D/aFPrTn5a/FkI53W8IgFxgsHgP0J/xc4OkdkC6VCl8woOl1bUqFxofZjMeAh3zGsNG9qrH4GiUQ1atj4vmGAQ=', 'supported_text': 'The method is founded on building in small steps with quick iterations and immediate feedback, and for coding specifically, involves writing 1-2 lines of code at a time and immediately showing the results.'}], [{'type': 'web_search_result_location', 'cited_text': 'An instance is your personal machine on which SolveIt runs. It’s a full virtual private server where you can install software, store files, and host a...', 'url': 'https://www.fast.ai/posts/2025-11-07-solveit-features.html', 'title': 'fast.ai - A Guide to Solveit Features', 'encrypted_index': 'EpEBCioIDxgCIiQ4ODk4YTFkYy0yMTNkLTRhNmYtOTljYi03ZTBlNTUzZDc0NWISDKpSJQLHkPYB7U5bahoMTyicEDCfNdGRSr2bIjBinugIV4MtfttAEqJFaw0RZiUs8JogZrWJJrRYmrJsMcPnVsnRN9SMP7SKJo2RNuMqFSWDHcLcmknK3SqHANVtfBysISEUrRgE', 'supported_text': 'The platform provides users with personal instances—full virtual private servers where they can install software, store files, and host applications.'}], [{'type': 'web_search_result_location', 'cited_text': 'It’s easier than ever to get started, but also easier than ever to let AI steer you into a situation where you’re overwhelmed by code you don’t unders...', 'url': 'https://www.answer.ai/posts/2025-10-01-solveit-full.html', 'title': 'Launching Solveit, the antidote to AI fatigue – Answer.AI', 'encrypted_index': 'EpEBCioIDxgCIiQ4ODk4YTFkYy0yMTNkLTRhNmYtOTljYi03ZTBlNTUzZDc0NWISDGgJQUovkBflUBq9tBoMPvcjHSbfagZ8o04aIjCSFbqI5pbxHTlex1P/fjFO1oyZcshLoQv0cEzV8eLzmyiu97QJ6tHdrRTQpIj0vLwqFZvp1AkoLBn6f5CQTX1cuRcxLciTohgE', 'supported_text': \"It serves as an antidote to AI fatigue, helping users avoid being overwhelmed by AI-generated code they don't understand, and has been tested with over 1000 preview users.\"}]], 'thinking_blocks': None, 'web_search_results': [{'type': 'web_search_tool_result', 'tool_use_id': 'srvtoolu_01VCLaffd3zxyZdvqNV1xkQ2', 'content': [{'type': 'web_search_result', 'title': 'Solve It With Code', 'url': 'https://solve.it.com/', 'encrypted_content': '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', 'page_age': None}, {'type': 'web_search_result', 'title': 'SolveIt | Full-Cycle Software Development Company', 'url': 'https://solveit.dev/', 'encrypted_content': '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', 'page_age': None}, {'type': 'web_search_result', 'title': 'fast.ai - A Guide to Solveit Features', 'url': 'https://www.fast.ai/posts/2025-11-07-solveit-features.html', 'encrypted_content': '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', 'page_age': 'November 7, 2025'}, {'type': 'web_search_result', 'title': '- Solve IT Solutions LLC', 'url': 'https://solve-it-sol.com/', 'encrypted_content': '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', 'page_age': 'April 26, 2024'}, {'type': 'web_search_result', 'title': 'Solveit – A course and platform for solving problems with code | Hacker News', 'url': 'https://news.ycombinator.com/item?id=45455719', 'encrypted_content': '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', 'page_age': 'October 15, 2025'}, {'type': 'web_search_result', 'title': 'Launching Solveit, the antidote to AI fatigue – Answer.AI', 'url': 'https://www.answer.ai/posts/2025-10-01-solveit-full.html', 'encrypted_content': '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', 'page_age': 'October 2, 2025'}, {'type': 'web_search_result', 'title': 'What Is Solveit? – johnowhitaker.dev', 'url': 'https://johnowhitaker.dev/posts/solveit.html', 'encrypted_content': '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', 'page_age': 'September 12, 2025'}, {'type': 'web_search_result', 'title': 'We have moved', 'url': 'https://solveit.fast.ai/', 'encrypted_content': 'EvcBCioIDxgCIiQ4ODk4YTFkYy0yMTNkLTRhNmYtOTljYi03ZTBlNTUzZDc0NWISDLaigGgZA/VH1wYfaxoM8qR6PAp/WyFQgRoJIjCt6gy67cBGT8sfwF2TDh+is5vdHLzpTxqJLmivxYWLP4VvAJ25/2UQoK52+0dPNKoqe2mLYeqM9Ec8KulygOWndH5tGGyW2xIZrDysi6bcs4muzIBcGcC6S232LKV6Zu+GtjTc3xIcsBc+zlF/KUs/DNOLecD8TXJgLpFQ7G5iEUbRQHDV3eWZHiS8jMWlIB+IFL/nIUKPuiIBhag74uqcBFzsPw5QAAkwXMBYmxgD', 'page_age': None}, {'type': 'web_search_result', 'title': 'SolveIT | Home', 'url': 'https://solveitservicesinc.com/', 'encrypted_content': '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', 'page_age': None}, {'type': 'web_search_result', 'title': 'SolveIT Managed Services', 'url': 'https://www.solveitms.com/', 'encrypted_content': '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', 'page_age': None}], 'caller': {'type': 'direct'}}]}))], usage=Usage(completion_tokens=386, prompt_tokens=12333, total_tokens=12719, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=386, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=12333, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), server_tool_use=ServerToolUse(web_search_requests=1, tool_search_requests=None), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None))"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='5 + 3 = **8**! Let me know if you have any other calculations! 😊', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=27, prompt_tokens=721, total_tokens=748, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=27, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
]
},
"metadata": {},
"output_type": "display_data"
- }
- ],
- "source": [
- "orig_completion = completion\n",
- "\n",
- "call_count = 0\n",
- "def patched_completion(*args, **kwargs):\n",
- " global call_count\n",
- " call_count += 1\n",
- " print(f\"Mock Call {call_count}\")\n",
- " if call_count < 3: return mk_pause_web_search()\n",
- " return orig_completion(*args, **kwargs)\n",
- "\n",
- "completion = patched_completion\n",
- "chat_pause = Chat('claude-sonnet-4-5', search='l')\n",
- "res = chat_pause(\"Search the web and tell me about Solveit in a paragraph\")\n",
- "print(f\"Total calls: {call_count}\")\n",
- "display(res)\n",
- "\n",
- "completion = orig_completion"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "e0062c45",
- "metadata": {},
- "source": [
- "Test next turn:"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "1fbb7a31",
- "metadata": {},
- "outputs": [],
- "source": [
- "test_eq(len(chat_pause.hist), 4)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "fb73be0e",
- "metadata": {},
- "outputs": [
+ },
{
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Mock Call 4\n"
- ]
+ "data": {
+ "text/markdown": [
+ "\n",
+ "\n",
+ "**openai/gpt-5.4:**\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(**openai/gpt-5.4:**)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
},
{
"data": {
"text/markdown": [
- "You asked me to search the web and tell you about Solveit in a paragraph.\n",
+ "8\n",
"\n",
"\n",
"\n",
"- id: `chatcmpl-xxx`\n",
- "- model: `claude-sonnet-4-5-20250929`\n",
+ "- model: `gpt-5.4`\n",
"- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=22, prompt_tokens=10334, total_tokens=10356, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=22, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=10334, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None)`\n",
+ "- usage: `Usage(completion_tokens=5, prompt_tokens=73, total_tokens=78, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None))`\n",
"\n",
" "
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-5-20250929', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='You asked me to search the web and tell you about Solveit in a paragraph.', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=22, prompt_tokens=10334, total_tokens=10356, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=22, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=10334, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None))"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gpt-5.4', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='8', role='assistant', tool_calls=None, function_call=None, provider_specific_fields=None))], usage=Usage(completion_tokens=5, prompt_tokens=73, total_tokens=78, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None)))"
]
},
- "execution_count": null,
"metadata": {},
- "output_type": "execute_result"
+ "output_type": "display_data"
}
],
"source": [
- "chat_pause('What did I just ask you about?')"
+ "for m in ms[1:]:\n",
+ " _sparams = litellm.get_model_info(m)['supported_openai_params']\n",
+ " if 'reasoning_effort' not in _sparams: continue\n",
+ " display(Markdown(f'**{m}:**'))\n",
+ " chat = Chat(m, tools=[simple_add])\n",
+ " res = chat(\"What's 5 + 3?\",think='l',return_all=True)\n",
+ " _display(*res)"
]
},
{
"cell_type": "markdown",
- "id": "bb3f1f79",
+ "id": "21fcaf1d",
"metadata": {},
"source": [
- "Workaround for https://github.com/BerriAI/litellm/issues/23047:"
+ "### Search"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "edc17903",
+ "id": "6fa66f10",
"metadata": {},
"outputs": [],
"source": [
"#| export\n",
- "import litellm.llms.anthropic.chat.transformation as _anth_t\n",
- "import litellm.litellm_core_utils.prompt_templates.factory as _fact"
+ "from litellm.litellm_core_utils.llm_cost_calc.tool_call_cost_tracking import StandardBuiltInToolCostTracking\n",
+ "from litellm.types.utils import ServerToolUse"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "e3278219",
+ "id": "beed6c09",
"metadata": {},
"outputs": [],
"source": [
"#| export\n",
- "# if '_orig_tp' not in globals(): _orig_tp = _anth_t.AnthropicConfig.transform_parsed_response\n",
- "# def _patched_tp(self, completion_response, raw_response, model_response, **kw):\n",
- "# r = _orig_tp(self, completion_response, raw_response, model_response, **kw)\n",
- "# msg = r.choices[0].message\n",
- "# psf = getattr(msg, 'provider_specific_fields', None) or {}\n",
- "# if len(psf.get('web_search_results') or []) > 1 and getattr(msg, 'thinking_blocks', None):\n",
- "# psf['_original_content'] = completion_response.get('content')\n",
- "# return r\n",
- "# \n",
- "# if '_orig_apt' not in globals(): _orig_apt = _fact.anthropic_messages_pt\n",
- "# def _patched_apt(messages, model, llm_provider):\n",
- "# for m in messages:\n",
- "# if not hasattr(m, 'get') or m.get('role') != 'assistant': continue\n",
- "# psf = m.get('provider_specific_fields')\n",
- "# if not isinstance(psf, dict) or not (oc := psf.get('_original_content')): continue\n",
- "# cc = next((b['cache_control'] for b in (m.get('content') or [])\n",
- "# if isinstance(b, dict) and 'cache_control' in b), None)\n",
- "# m['content'],m['thinking_blocks'] = oc,None\n",
- "# m['tool_calls'] = [tc for tc in (m.get('tool_calls') or [])\n",
- "# if not (getattr(tc, 'id', '') or tc.get('id', '')).startswith('srvtoolu_')] or None\n",
- "# for k in ('web_search_results', 'tool_results'): psf.pop(k, None)\n",
- "# if cc: oc[-1]['cache_control'] = cc\n",
- "# return _orig_apt(messages, model, llm_provider)\n",
- "# \n",
- "# _anth_t.AnthropicConfig.transform_parsed_response = _patched_tp\n",
- "# _fact.anthropic_messages_pt = _patched_apt"
+ "# Workaround for https://github.com/BerriAI/litellm/issues/26153\n",
+ "@patch_to(StandardBuiltInToolCostTracking, cls_method=True)\n",
+ "def get_cost_for_built_in_tools(cls, model, response_object=None, usage=None, custom_llm_provider=None, standard_built_in_tools_params=None):\n",
+ " if usage is not None and isinstance(getattr(usage, 'server_tool_use', None), dict):\n",
+ " usage.server_tool_use = ServerToolUse(**usage.server_tool_use)\n",
+ " return cls._orig_get_cost_for_built_in_tools(model, response_object=response_object, usage=usage, custom_llm_provider=custom_llm_provider, standard_built_in_tools_params=standard_built_in_tools_params)"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "d2bdb7c7",
- "metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Turn 1: thinking=True, tcs=[ChatCompletionMessageToolCall(index=2, function=Function(arguments='{\"query\": \"fast.ai latest news 2026\"}', name='web_search'), id='srvtoolu_01JZtcwb1b8xxxBMsfDE6tBt', type='function'), ChatCompletionMessageToolCall(index=3, function=Function(arguments='{\"query\": \"answer.ai latest news 2026\"}', name='web_search'), id='srvtoolu_01RgBPcesVTnc25Yc16XibJ5', type='function')]\n",
- "Turn 2 OK\n"
- ]
- }
- ],
- "source": [
- "m = 'claude-sonnet-4-6'\n",
- "msgs = [{'role':'user','content':\"Search web for latest news about fast.ai and answer.ai.\"}]\n",
- "r = completion(m, msgs, web_search_options={\"search_context_size\":\"low\"}, reasoning_effort='low')\n",
- "m1 = r.choices[0].message\n",
- "print(f\"Turn 1: thinking={bool(m1.thinking_blocks)}, tcs={m1.tool_calls}\")\n",
- "\n",
- "msgs.append(m1)\n",
- "msgs.append({'role':'user','content':'And search for news about solveit.'})\n",
- "r2 = completion(m, msgs, web_search_options={\"search_context_size\":\"low\"}, reasoning_effort='low')\n",
- "print(\"Turn 2 OK\")"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "537c93a5",
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/markdown": [
- "123 + 456 = **579**. ([fast.ai](https://www.fast.ai/?utm_source=openai))\n",
- "\n",
- "I searched for **today’s fast.ai news** and didn’t find any clearly dated **Monday, April 27, 2026** news item from fast.ai itself. The fastest/most reliable signal I found was the official fast.ai homepage, which was crawled today and shows the most recent published post as **“I Don’t Want a Learning Dashboard for My Child”** from **February 17, 2026**. It also lists **“Breaking the Spell of Vibe Coding”** from **January 28, 2026** and **“How To Use AI for the Ancient Art of Close Reading”** from **January 21, 2026**. ([fast.ai](https://www.fast.ai/?utm_source=openai))\n",
- "\n",
- "So the short answer is: **no obvious same-day fast.ai news showed up in web results today**; the latest visible official updates I found are those early-2026 blog posts. ([fast.ai](https://www.fast.ai/?utm_source=openai))\n",
- "\n",
- "If you want, I can do a second pass focused specifically on:\n",
- "1. **official fast.ai/Answer.AI announcements**, \n",
- "2. **Jeremy Howard news/interviews**, or \n",
- "3. **GitHub/forum activity from the last 7 days**.\n",
- "\n",
- "\n",
- "\n",
- "- id: `chatcmpl-xxx`\n",
- "- model: `gpt-5.4`\n",
- "- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=741, prompt_tokens=18567, total_tokens=19308, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=409, rejected_prediction_tokens=None, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=4608, text_tokens=None, image_tokens=None, video_tokens=None))`\n",
- "\n",
- " "
- ],
- "text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gpt-5.4', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='123 + 456 = **579**. ([fast.ai](https://www.fast.ai/?utm_source=openai))\\n\\nI searched for **today’s fast.ai news** and didn’t find any clearly dated **Monday, April 27, 2026** news item from fast.ai itself. The fastest/most reliable signal I found was the official fast.ai homepage, which was crawled today and shows the most recent published post as **“I Don’t Want a Learning Dashboard for My Child”** from **February 17, 2026**. It also lists **“Breaking the Spell of Vibe Coding”** from **January 28, 2026** and **“How To Use AI for the Ancient Art of Close Reading”** from **January 21, 2026**. ([fast.ai](https://www.fast.ai/?utm_source=openai))\\n\\nSo the short answer is: **no obvious same-day fast.ai news showed up in web results today**; the latest visible official updates I found are those early-2026 blog posts. ([fast.ai](https://www.fast.ai/?utm_source=openai))\\n\\nIf you want, I can do a second pass focused specifically on:\\n1. **official fast.ai/Answer.AI announcements**, \\n2. **Jeremy Howard news/interviews**, or \\n3. **GitHub/forum activity from the last 7 days**.', role='assistant', tool_calls=None, function_call=None, reasoning_content='', reasoning_items=[{'type': 'reasoning', 'id': 'rs_01c02d6998ad23ac0069eebd47e01c819398adfb4d16dfc8ae', 'encrypted_content': None, 'summary': []}], provider_specific_fields=None, annotations=[{'type': 'url_citation', 'end_index': 72, 'start_index': 21, 'title': 'fast.ai - fast.ai—Making neural nets uncool again', 'url': 'https://www.fast.ai/?utm_source=openai'}, {'type': 'url_citation', 'end_index': 650, 'start_index': 599, 'title': 'fast.ai - fast.ai—Making neural nets uncool again', 'url': 'https://www.fast.ai/?utm_source=openai'}, {'type': 'url_citation', 'end_index': 873, 'start_index': 822, 'title': 'fast.ai - fast.ai—Making neural nets uncool again', 'url': 'https://www.fast.ai/?utm_source=openai'}]))], usage=Usage(completion_tokens=741, prompt_tokens=18567, total_tokens=19308, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=409, rejected_prediction_tokens=None, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=4608, text_tokens=None, image_tokens=None, video_tokens=None)))"
- ]
- },
- "execution_count": null,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "chat = Chat(gpt54, tools=[simple_add])\n",
- "chat(\"What's 123+456? Then search the web for today's news about fast.ai.\", think='l', search='l')"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "d8cce2d9",
- "metadata": {},
- "source": [
- "### Multi tool calling"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "bdb42380",
- "metadata": {},
- "source": [
- "We can let the model call multiple tools in sequence using the `max_steps` parameter."
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "662f7aa3",
+ "id": "afd9a499",
"metadata": {},
"outputs": [
{
@@ -6581,22 +7060,24 @@
{
"data": {
"text/markdown": [
+ "Otters are semiaquatic, carnivorous mammals belonging to the weasel family. There are 13 species worldwide, inhabiting both freshwater rivers and marine environments. \n",
"\n",
- "\n",
- "🔧 simple_add({\"a\": 5, \"b\": 3})\n",
- "\n",
+ "Here is a brief overview of their key characteristics:\n",
+ "* **Appearance:** They have long, sleek bodies, webbed feet for swimming, and incredibly dense fur—sea otters actually have the thickest fur of any animal on Earth to insulate them in cold waters. \n",
+ "* **Diet:** They primarily eat fish, crabs, clams, and amphibians. \n",
+ "* **Behavior:** Otters are highly intelligent and famously playful. Sea otters, for example, sleep floating on their backs and are one of the few mammals that use tools, often using rocks to smash open shellfish on their bellies.\n",
"\n",
"\n",
"\n",
"- id: `chatcmpl-xxx`\n",
"- model: `gemini-3-pro-preview`\n",
- "- finish_reason: `tool_calls`\n",
- "- usage: `Usage(completion_tokens=174, prompt_tokens=93, total_tokens=267, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=156, rejected_prediction_tokens=None, text_tokens=18, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=93, image_tokens=None, video_tokens=None), cache_read_input_tokens=None)`\n",
+ "- finish_reason: `stop`\n",
+ "- usage: `Usage(completion_tokens=462, prompt_tokens=192, total_tokens=654, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=253, rejected_prediction_tokens=None, text_tokens=209, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=192, image_tokens=None, video_tokens=None, web_search_requests=1))`\n",
"\n",
" "
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-pro-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=None, role='assistant', tool_calls=[{'index': 0, 'provider_specific_fields': {'thought_signature': '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'}, 'function': {'arguments': '{\"a\": 5, \"b\": 3}', 'name': 'simple_add'}, 'id': 'call_f153162026964e02a1ad06738830__thought__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', 'type': 'function'}], function_call=None, images=[], thinking_blocks=[], provider_specific_fields={'thought_signatures': ['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']}))], usage=Usage(completion_tokens=174, prompt_tokens=93, total_tokens=267, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=156, rejected_prediction_tokens=None, text_tokens=18, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=93, image_tokens=None, video_tokens=None), cache_read_input_tokens=None), vertex_ai_grounding_metadata=[], vertex_ai_url_context_metadata=[], vertex_ai_safety_results=[], vertex_ai_citation_metadata=[], service_tier='default')"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-pro-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='Otters are semiaquatic, carnivorous mammals belonging to the weasel family. There are 13 species worldwide, inhabiting both freshwater rivers and marine environments. \\n\\nHere is a brief overview of their key characteristics:\\n* **Appearance:** They have long, sleek bodies, webbed feet for swimming, and incredibly dense fur—sea otters actually have the thickest fur of any animal on Earth to insulate them in cold waters. \\n* **Diet:** They primarily eat fish, crabs, clams, and amphibians. \\n* **Behavior:** Otters are highly intelligent and famously playful. Sea otters, for example, sleep floating on their backs and are one of the few mammals that use tools, often using rocks to smash open shellfish on their bellies.', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'thought_signatures': ['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', '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', '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'], 'server_side_tool_invocations': [{'tool_type': 'GOOGLE_SEARCH_WEB', 'id': 'x21jhl2n', 'args': {'queries': ['otters']}, 'thought_signature': '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', 'response': {'search_suggestions': '\\n\\n
\\n
\\n \\n \\n \\n \\n \\n
\\n \\n \\n \\n \\n \\n \\n
\\n
\\n
\\n
\n",
+ "\n",
+ "**gemini/gemini-3-flash-preview:**\n",
+ "\n",
+ "
"
+ ],
"text/plain": [
- "'8'"
+ "Markdown(**gemini/gemini-3-flash-preview:**)"
]
},
"metadata": {},
@@ -6614,22 +7102,26 @@
{
"data": {
"text/markdown": [
+ "Otters are highly intelligent, playful, semi-aquatic mammals belonging to the weasel family (**Mustelidae**). There are **13 extant species** found on every continent except Australia and Antarctica.\n",
"\n",
- "\n",
- "🔧 simple_add({\"b\": 7, \"a\": 8})\n",
- "\n",
+ "### Key Facts\n",
+ "* **Physical Traits:** They have long, streamlined bodies, powerful webbed feet for swimming, and extremely dense, water-repellent fur that keeps them warm. They are the only members of the weasel family that are serious swimmers.\n",
+ "* **Habitats:** Otters live in both freshwater (rivers, lakes, and wetlands) and saltwater (coastal marine environments). While river otters spend significant time on land, sea otters live almost exclusively in the water.\n",
+ "* **Behavior:** They are famous for their \"playful\" nature—often seen sliding down mud banks or playing with stones. They are carnivorous, primarily eating fish, crustaceans, and mollusks. Some species, like sea otters, use rocks as tools to crack open shells.\n",
+ "* **Size Range:** They vary greatly in size, from the **Asian small-clawed otter** (roughly 3–10 lbs) to the **Giant Otter** of South America, which can reach up to 6 feet in length, and the **Sea Otter**, which can weigh up to 100 lbs.\n",
+ "* **Conservation:** Many species are currently threatened or endangered due to habitat loss, pollution, and historical hunting for their thick pelts.\n",
"\n",
"\n",
"\n",
"- id: `chatcmpl-xxx`\n",
- "- model: `gemini-3-pro-preview`\n",
- "- finish_reason: `tool_calls`\n",
- "- usage: `Usage(completion_tokens=42, prompt_tokens=281, total_tokens=323, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=24, rejected_prediction_tokens=None, text_tokens=18, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=281, image_tokens=None, video_tokens=None), cache_read_input_tokens=None)`\n",
+ "- model: `gemini-3-flash-preview`\n",
+ "- finish_reason: `stop`\n",
+ "- usage: `Usage(completion_tokens=683, prompt_tokens=76, total_tokens=759, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=380, rejected_prediction_tokens=None, text_tokens=303, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=76, image_tokens=None, video_tokens=None, web_search_requests=1))`\n",
"\n",
" "
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-pro-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=None, role='assistant', tool_calls=[{'index': 0, 'provider_specific_fields': {'thought_signature': 'EmgKZgEMOdbHRr0x5vG/DkqBSn3ZpVY87OIU84H2M5vffH769pzWIU3Vqqj+ztRPpwan7T2hk6uRpn1dsByPvlHh66uMQDRxwS1LcmmefXopD97inFuFIf6Ljq+ID2AUJQjxnup+yOoYLg=='}, 'function': {'arguments': '{\"b\": 7, \"a\": 8}', 'name': 'simple_add'}, 'id': 'call_d98449c9595940b998e4a318245d__thought__EmgKZgEMOdbHRr0x5vG/DkqBSn3ZpVY87OIU84H2M5vffH769pzWIU3Vqqj+ztRPpwan7T2hk6uRpn1dsByPvlHh66uMQDRxwS1LcmmefXopD97inFuFIf6Ljq+ID2AUJQjxnup+yOoYLg==', 'type': 'function'}], function_call=None, images=[], thinking_blocks=[], provider_specific_fields={'thought_signatures': ['EmgKZgEMOdbHRr0x5vG/DkqBSn3ZpVY87OIU84H2M5vffH769pzWIU3Vqqj+ztRPpwan7T2hk6uRpn1dsByPvlHh66uMQDRxwS1LcmmefXopD97inFuFIf6Ljq+ID2AUJQjxnup+yOoYLg==']}))], usage=Usage(completion_tokens=42, prompt_tokens=281, total_tokens=323, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=24, rejected_prediction_tokens=None, text_tokens=18, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=281, image_tokens=None, video_tokens=None), cache_read_input_tokens=None), vertex_ai_grounding_metadata=[], vertex_ai_url_context_metadata=[], vertex_ai_safety_results=[], vertex_ai_citation_metadata=[], service_tier='default')"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-flash-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='Otters are highly intelligent, playful, semi-aquatic mammals belonging to the weasel family (**Mustelidae**). There are **13 extant species** found on every continent except Australia and Antarctica.\\n\\n### Key Facts\\n* **Physical Traits:** They have long, streamlined bodies, powerful webbed feet for swimming, and extremely dense, water-repellent fur that keeps them warm. They are the only members of the weasel family that are serious swimmers.\\n* **Habitats:** Otters live in both freshwater (rivers, lakes, and wetlands) and saltwater (coastal marine environments). While river otters spend significant time on land, sea otters live almost exclusively in the water.\\n* **Behavior:** They are famous for their \"playful\" nature—often seen sliding down mud banks or playing with stones. They are carnivorous, primarily eating fish, crustaceans, and mollusks. Some species, like sea otters, use rocks as tools to crack open shells.\\n* **Size Range:** They vary greatly in size, from the **Asian small-clawed otter** (roughly 3–10 lbs) to the **Giant Otter** of South America, which can reach up to 6 feet in length, and the **Sea Otter**, which can weigh up to 100 lbs.\\n* **Conservation:** Many species are currently threatened or endangered due to habitat loss, pollution, and historical hunting for their thick pelts.', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'thought_signatures': ['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', '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', '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'], 'server_side_tool_invocations': [{'tool_type': 'GOOGLE_SEARCH_WEB', 'id': 'nsdqfchu', 'args': {'queries': ['brief summary about otters physical characteristics habitat behavior species']}, 'thought_signature': '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', 'response': {'search_suggestions': '\\n\\n
\\n
\\n \\n \\n \\n \\n \\n
\\n \\n \\n \\n \\n \\n \\n
\\n
\\n
\\n
\n",
+ "\n",
+ "**claude-sonnet-4-6:**\n",
+ "\n",
+ "
"
+ ],
"text/plain": [
- "'15'"
+ "Markdown(**claude-sonnet-4-6:**)"
]
},
"metadata": {},
@@ -6647,134 +7146,397 @@
{
"data": {
"text/markdown": [
+ "Here's a brief overview of otters:\n",
"\n",
+ "**What they are:** [*](https://en.wikipedia.org/wiki/Otter \"Otter - Wikipedia\") Otters are carnivorous mammals in the subfamily Lutrinae, and all 14 extant species are semiaquatic, living in both freshwater and marine environments. [*](https://www.nationalgeographic.com/animals/mammals/facts/otters-1 \"Otters, facts and information | National Geographic\") They are found on every continent except Australia and Antarctica.\n",
"\n",
- "🔧 simple_add({\"a\": 15, \"b\": 11})\n",
+ "**Physical traits:** [*](https://en.wikipedia.org/wiki/Otter \"Otter - Wikipedia\") Otters are distinguished by their long, slim bodies, powerful webbed feet for swimming, and dense fur, which keeps them warm and buoyant in water. [*](https://www.nationalgeographic.com/animals/mammals/facts/otters-1 \"Otters, facts and information | National Geographic\") They have the densest fur of any animal — as many as a million hairs per square inch in places.\n",
+ "\n",
+ "**Diet & tools:** [*](https://www.nationalgeographic.com/animals/mammals/facts/otters-1 \"Otters, facts and information | National Geographic\") All otters are expert hunters that eat fish, crustaceans, and other critters. [*](https://www.nationalgeographic.com/animals/mammals/facts/otters-1 \"Otters, facts and information | National Geographic\") Sea otters have an ingenious method to open shellfish — a sea otter will float on its back, place a rock on its chest, then smash the mollusk down on it until it breaks open.\n",
+ "\n",
+ "**Behavior:** [*](https://en.wikipedia.org/wiki/Otter \"Otter - Wikipedia\") They are playful animals, engaging in activities like sliding into water on natural slides and playing with stones. [*](https://www.nationalgeographic.com/animals/mammals/facts/otters-1 \"Otters, facts and information | National Geographic\") When it's time to nap, sea otters entangle themselves in kelp so they don't float away, and they sometimes intertwine their feet with another sea otter to stay together.\n",
+ "\n",
+ "**Lifespan & young:** [*](https://en.wikipedia.org/wiki/Otter \"Otter - Wikipedia\") They can live up to 16 years, with their diet mainly consisting of fish and sometimes frogs, birds, or shellfish, depending on the species. [*](https://www.doi.gov/blog/12-facts-about-otters-sea-otter-awareness-week \"12 Facts About Otters for Sea Otter Awareness Week | U.S. Department of the Interior\") A newborn pup needs constant attention and will stay with its mother for six months until it develops survival skills.\n",
+ "\n",
+ "**Conservation:** [*](https://www.nationalgeographic.com/animals/mammals/facts/otters-1 \"Otters, facts and information | National Geographic\") Otters and their relatives were once hunted extensively for their fur, many to the point of near extinction, and despite regulations designed to protect them, many species remain at risk from pollution and habitat loss.\n",
+ "\n",
+ "🔧 web_search({\"query\": \"otters facts overview\"})\n",
"\n",
"\n",
"\n",
"\n",
"- id: `chatcmpl-xxx`\n",
- "- model: `gemini-3-pro-preview`\n",
- "- finish_reason: `tool_calls`\n",
- "- usage: `Usage(completion_tokens=46, prompt_tokens=338, total_tokens=384, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=26, rejected_prediction_tokens=None, text_tokens=20, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=338, image_tokens=None, video_tokens=None), cache_read_input_tokens=None)`\n",
+ "- model: `claude-sonnet-4-6`\n",
+ "- finish_reason: `stop`\n",
+ "- usage: `Usage(completion_tokens=627, prompt_tokens=17556, total_tokens=18183, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=627, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0), server_tool_use=ServerToolUse(web_search_requests=1, tool_search_requests=None), cache_creation_input_tokens=0, cache_read_input_tokens=0)`\n",
"\n",
" "
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-pro-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=None, role='assistant', tool_calls=[{'index': 0, 'provider_specific_fields': {'thought_signature': 'EmoKaAEMOdbHDR1RCph7o/YOes1BfT+NIMvXTb8bVW1jFZ65dq1EWnui1mOA2QX5LRzNNTUzDCHMckhne3kSvdaNWCqaFRLGYggV8yTAdZufWGaWNXsxbk9YI3EbCfELckuM4mkAm4TqhNFI'}, 'function': {'arguments': '{\"a\": 15, \"b\": 11}', 'name': 'simple_add'}, 'id': 'call_904c48ebf3934e8ea51b00544f0d__thought__EmoKaAEMOdbHDR1RCph7o/YOes1BfT+NIMvXTb8bVW1jFZ65dq1EWnui1mOA2QX5LRzNNTUzDCHMckhne3kSvdaNWCqaFRLGYggV8yTAdZufWGaWNXsxbk9YI3EbCfELckuM4mkAm4TqhNFI', 'type': 'function'}], function_call=None, images=[], thinking_blocks=[], provider_specific_fields={'thought_signatures': ['EmoKaAEMOdbHDR1RCph7o/YOes1BfT+NIMvXTb8bVW1jFZ65dq1EWnui1mOA2QX5LRzNNTUzDCHMckhne3kSvdaNWCqaFRLGYggV8yTAdZufWGaWNXsxbk9YI3EbCfELckuM4mkAm4TqhNFI']}))], usage=Usage(completion_tokens=46, prompt_tokens=338, total_tokens=384, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=26, rejected_prediction_tokens=None, text_tokens=20, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=338, image_tokens=None, video_tokens=None), cache_read_input_tokens=None), vertex_ai_grounding_metadata=[], vertex_ai_url_context_metadata=[], vertex_ai_safety_results=[], vertex_ai_citation_metadata=[], service_tier='default')"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "text/plain": [
- "'26'"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='Here\\'s a brief overview of otters:\\n\\n**What they are:** [*](https://en.wikipedia.org/wiki/Otter \"Otter - Wikipedia\") Otters are carnivorous mammals in the subfamily Lutrinae, and all 14 extant species are semiaquatic, living in both freshwater and marine environments. [*](https://www.nationalgeographic.com/animals/mammals/facts/otters-1 \"Otters, facts and information | National Geographic\") They are found on every continent except Australia and Antarctica.\\n\\n**Physical traits:** [*](https://en.wikipedia.org/wiki/Otter \"Otter - Wikipedia\") Otters are distinguished by their long, slim bodies, powerful webbed feet for swimming, and dense fur, which keeps them warm and buoyant in water. [*](https://www.nationalgeographic.com/animals/mammals/facts/otters-1 \"Otters, facts and information | National Geographic\") They have the densest fur of any animal — as many as a million hairs per square inch in places.\\n\\n**Diet & tools:** [*](https://www.nationalgeographic.com/animals/mammals/facts/otters-1 \"Otters, facts and information | National Geographic\") All otters are expert hunters that eat fish, crustaceans, and other critters. [*](https://www.nationalgeographic.com/animals/mammals/facts/otters-1 \"Otters, facts and information | National Geographic\") Sea otters have an ingenious method to open shellfish — a sea otter will float on its back, place a rock on its chest, then smash the mollusk down on it until it breaks open.\\n\\n**Behavior:** [*](https://en.wikipedia.org/wiki/Otter \"Otter - Wikipedia\") They are playful animals, engaging in activities like sliding into water on natural slides and playing with stones. [*](https://www.nationalgeographic.com/animals/mammals/facts/otters-1 \"Otters, facts and information | National Geographic\") When it\\'s time to nap, sea otters entangle themselves in kelp so they don\\'t float away, and they sometimes intertwine their feet with another sea otter to stay together.\\n\\n**Lifespan & young:** [*](https://en.wikipedia.org/wiki/Otter \"Otter - Wikipedia\") They can live up to 16 years, with their diet mainly consisting of fish and sometimes frogs, birds, or shellfish, depending on the species. [*](https://www.doi.gov/blog/12-facts-about-otters-sea-otter-awareness-week \"12 Facts About Otters for Sea Otter Awareness Week | U.S. Department of the Interior\") A newborn pup needs constant attention and will stay with its mother for six months until it develops survival skills.\\n\\n**Conservation:** [*](https://www.nationalgeographic.com/animals/mammals/facts/otters-1 \"Otters, facts and information | National Geographic\") Otters and their relatives were once hunted extensively for their fur, many to the point of near extinction, and despite regulations designed to protect them, many species remain at risk from pollution and habitat loss.', role='assistant', tool_calls=[ChatCompletionMessageToolCall(function=Function(arguments='{\"query\": \"otters facts overview\"}', name='web_search'), id='srvtoolu_016FZfqBdBegL6dwXMvCyBHd', type='function')], function_call=None, provider_specific_fields={'web_search_results': [{'type': 'web_search_tool_result', 'tool_use_id': 'srvtoolu_016FZfqBdBegL6dwXMvCyBHd', 'content': [{'type': 'web_search_result', 'title': '12 Facts About Otters for Sea Otter Awareness Week | U.S. Department of the Interior', 'url': 'https://www.doi.gov/blog/12-facts-about-otters-sea-otter-awareness-week', 'encrypted_content': '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', 'page_age': None}, {'type': 'web_search_result', 'title': 'Otter - Wikipedia', 'url': 'https://en.wikipedia.org/wiki/Otter', 'encrypted_content': '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', 'page_age': '1 month ago'}, {'type': 'web_search_result', 'title': 'Otter | Description, Habitat, Species, Distribution, Food, & Facts | Britannica', 'url': 'https://www.britannica.com/animal/otter', 'encrypted_content': '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', 'page_age': '3 weeks ago'}, {'type': 'web_search_result', 'title': 'Otters, facts and information | National Geographic', 'url': 'https://www.nationalgeographic.com/animals/mammals/facts/otters-1', 'encrypted_content': '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', 'page_age': 'May 4, 2021'}, {'type': 'web_search_result', 'title': 'Otter Fact Sheet | Blog | Nature | PBS', 'url': 'https://www.pbs.org/wnet/nature/blog/otter-fact-sheet/', 'encrypted_content': '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', 'page_age': 'February 11, 2026'}, {'type': 'web_search_result', 'title': 'River Otter | National Geographic Kids', 'url': 'https://kids.nationalgeographic.com/animals/mammals/facts/river-otter', 'encrypted_content': 'ErcNCioIDxgCIiQ4ODk4YTFkYy0yMTNkLTRhNmYtOTljYi03ZTBlNTUzZDc0NWISDN+yI4FF7fPFfXXgLhoMFMLPvf3j9QYE5xWhIjBYnsfCSEijGbLQQulQB6nDkUIMz20ac5Xwzh25KBKrYuT7x6Lt/y9Pt5578jV6vZkqugzOb1DKGWfXf0CyRQXkSja6gXnJDQEQ4X0o75HW8swf043LOLDeo4AjJZMutFDK3Ia7jZtBemGejuTfAqlFlkF8OXx7+AyvEOg1Zz190ExNPANzr98Zzg+8dfm6CC6jexVrZPP2DbeXboh/+FknXw3gaGk19A1RzxRtXIW9JI95mGqgZR+RwI61Qg2uZbaO6wPhW3TUkD/c2mqHUl+xSHzONMUEabpNNCm+6b8OBm3CT+SLVegVlUo0mgKqlIHU1ouk4Ms1Acj0m0JC8I++6EaPQwPvBuOrJnpcFk8XSbG5f+hEOKQffivP2byVswiKHuL6KvmZh8tF8ZwIkNmCHqHlXxaLH6pvAk/vlgaO7OAiv2VndMOglhjAgu8KZmqh3uNAZHuBkuQqGVIEAXqUogKhXpco8L2DH+00fUTzG9WrJri41emXYoxgJvozX/u0TO8RtpXtDcqDTSJlCzOeH8wBzeervcfwzPYBFWKzN6e9jg+nihyMc7NfxJdWao55xILPpAsEY0hBxZ05Ew4xpsq25I+Cs7eIP10FM7SCF1aUk69TiCtX21jiYv68o2h59miNoIPY4ZetdhqOE2axRfxquZ/pXRGJ1KyqVdSbwBemavlVkyRlm9G61Fn+Xip6eOG0mCpSU6hp4YekXz2Z3FO5qfNYcm4LfJzbEzpTkU4504EgsH7m0EU+Fu436Tt4Slk9LaP+zOTJXOVRAC7T/ikBt3T1Anoa9PILchocqZ6WfsqyXza+KaETtIjItSopYZsdvNr5MHNjbXcUhfxRsTDItVhU3qr5mGOprniNZaMQa78PZriSK/+lsViAVl7B+lcchXL2+IdMPBzoUUr7nEK8zgAkgDRsSGeqKCj5DWJOGXZrwkwSJ3kj0FcYsG1LnFAf4/0av3BVhkP5wxjKmuDEvAmBcpI6MR7zPzGhTkIhP2mthvGdbwjrlEBqo++rDutuTnDzvnQJeZXJlMfxm1dqogqvOtnf/JfbVQ/yZ0x2tnHouwEiMRN1Ga/F1cFrokKdTmQfJPoj3QcQzhaZjNmowqh5qMs/yE+Xs3UllEg/0SktylYvSbzTnMgzBBoOGkOZHBQXgw7oUfxxzTPAuvexSlwQzbvBiubcY5jQLPB68EwgsyZueXkqZtwdFCzrKn3Py4J50mMe9Ru78k5LhXLGblgtCnKlMUow8Ug/YBYgmgX1C+gz4CuLuDFigInGN75XkIxRBqhXodBtaUJWFUQfbs9V13adQEXOCM7AAey9pTFaem8yqaHhDZzfBySAgM2n2ECNpeCZw59ZWPvjAyS1YUYMVf8WFFF+TnW5vAaQoEpPZwE1hj26M4tayPadJEjAWvYnkhQoRk1Q3O7Q+dmw5RrHyGiIgdPraAy9TQEVNKHgYKi/uZPcedmBoOk44u4TIqNJRCKmX05PMst/WV5EWY1qka53rHfkrX0fUdJhi4JHiJ5iXUP9zyek56dot098jaOCZoDZ/u/tXAj2wM4hg1Murg9WLOgB7U10LDtu0PUQCYv7/oO9djj8oWX+IIyrls0FeXyR3wtRA8pb7gjXoe/Q1M1XRdHgC6Z5vp3avcEoxkM4tpK/+X+hzowiMYPLP1UHMN983R8rGxnOaSrr7NE+bRJ+t9rN9rsSx+bT4uD1XYorvVBZi2m+TfWUP0FahYTm8mJ+RenfyPyBFh3skTL/xhkjDDxG6VdFzqngpYvoBPCaeurFSTioqr+FOex1OXkPg4HjgZbGmiUTQrf77h54UGKq5MHPJ5xX8zgrawKwgDpBULI0v8PFV/Lsd3fKNXQ7kVj52Hw5nubFS1t9kVtY1MsDQ8NzLNMWQ7l8d9AcnFIb4EFBWj+oI3JDN9yVk9e1B8uqcqL8msYBuWesM9Y/qWxENNDMuXI44QmSBwu7mLIO/3PG4nmFTP0YK/SPNnhyOaM52wRP6DIK1wgOjExmNytaK5zYJ2b1GM3VM5Pyeo/Ds023B8qUBs6vTxEiXO4+tVHOePe8sQoPWGDMExdlE77f7hqqEsBs/8aWB4nL/rQQXUS6N/8/BoV94lAmdGfjRKUClbRW44mAYIZDG0HsPsrlhBU4B2Nckg9Vm1tszudTPW0vXpSdoL/BDh26HbWmNUcaR4D2GAM=', 'page_age': 'October 6, 2023'}, {'type': 'web_search_result', 'title': 'Seven Quick Facts About River Otters - National Forest Foundation', 'url': 'https://www.nationalforests.org/article/seven-quick-facts-about-river-otters/', 'encrypted_content': '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', 'page_age': 'November 18, 2014'}, {'type': 'web_search_result', 'title': 'Otter Fun Facts! - WonderLab Museum', 'url': 'https://wonderlab.org/otter-fun-facts/', 'encrypted_content': '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', 'page_age': 'February 20, 2024'}, {'type': 'web_search_result', 'title': '18 Otter Facts From the Playful to Slightly Bizarre - TRVST', 'url': 'https://www.trvst.world/biodiversity/otter-facts/', 'encrypted_content': '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', 'page_age': 'December 7, 2024'}, {'type': 'web_search_result', 'title': 'Otters • Chesapeake Bay Foundation', 'url': 'https://www.cbf.org/nature/otters/', 'encrypted_content': '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', 'page_age': 'December 3, 2025'}], 'caller': {'type': 'direct'}}], 'citation': {'type': 'web_search_result_location', 'cited_text': 'Otters and their mustelid relatives were once hunted extensively for their fur, many to the point of near extinction. Despite regulations designed to ...', 'url': 'https://www.nationalgeographic.com/animals/mammals/facts/otters-1', 'title': 'Otters, facts and information | National Geographic', 'encrypted_index': 'EpMBCioIDxgCIiQ4ODk4YTFkYy0yMTNkLTRhNmYtOTljYi03ZTBlNTUzZDc0NWISDMIIEh2AaP0UlVZkzxoMBitXGNpXydPJqYXPIjAjhZaLLwG4+yA8RTTcdSjDcSYHs1pSCDehQsV7gd4sPqBg18Ly1CIzqXkBNHEWlbkqFz/hb6VQ/DCCzxa1ZJwbufPU45VCLY8zGAQ='}}))], usage=Usage(completion_tokens=627, prompt_tokens=17556, total_tokens=18183, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=627, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0), server_tool_use=ServerToolUse(web_search_requests=1, tool_search_requests=None), cache_creation_input_tokens=0, cache_read_input_tokens=0))"
]
},
"metadata": {},
"output_type": "display_data"
- },
+ }
+ ],
+ "source": [
+ "for m in ms[:-1]:\n",
+ " display(Markdown(f'**{m}:**'))\n",
+ " chat = Chat(m)\n",
+ " res = chat(\"Search the web and tell me very briefly about otters\", search='l', stream=True)\n",
+ " for o in res:\n",
+ " if isinstance(o, ModelResponse): sleep(0.01); display(o)\n",
+ " else: pass"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "7dbab35b",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "m = 'claude-sonnet-4-6'"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "18f95e5e",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "def mk_pause_web_search():\n",
+ " srv_tc = mk_tc(\"web_search\", query=\"Solveit Answer.AI\", tcid=random_tool_id().replace('toolu_', 'srvtoolu_'))\n",
+ " pause_msg = mk_tc_req(\"Let me search for that information:\", [srv_tc])\n",
+ " return ModelResponse(choices=[Choices.model_construct(finish_reason=\"pause_turn\", index=0, message=pause_msg)])"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "f7c25d76",
+ "metadata": {},
+ "outputs": [
{
"data": {
"text/markdown": [
- "The result of ((5 + 3) + 7) + 11 is 26.\n",
+ "Let me search for that information:\n",
+ "\n",
+ "🔧 web_search({\"query\": \"Solveit Answer.AI\"})\n",
+ "\n",
"\n",
"\n",
"\n",
"- id: `chatcmpl-xxx`\n",
- "- model: `gemini-3-pro-preview`\n",
- "- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=46, prompt_tokens=399, total_tokens=445, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=24, rejected_prediction_tokens=None, text_tokens=22, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=399, image_tokens=None, video_tokens=None), cache_read_input_tokens=None)`\n",
+ "- model: `None`\n",
+ "- finish_reason: `pause_turn`\n",
"\n",
" "
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-pro-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='The result of ((5 + 3) + 7) + 11 is 26.', role='assistant', tool_calls=None, function_call=None, images=[], thinking_blocks=[], provider_specific_fields={'thought_signatures': ['EokBCoYBAQw51sdzis/bOYhfim+7lao+teR/v50sqVFzqb7DLCt2reNskgMZBvD7xwjS4aDD+Fumrb5uL4Sy233VRWuW7Bz4XlNXbuhwv89fl2wh3xIk42pW3lPOJN4HdbeG6zBcXtePy7Ys8sgUFLdsakr4Tvoemvn4COKUy1d09eKP5nuBRAp0m48=']}))], usage=Usage(completion_tokens=46, prompt_tokens=399, total_tokens=445, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=24, rejected_prediction_tokens=None, text_tokens=22, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=399, image_tokens=None, video_tokens=None), cache_read_input_tokens=None), vertex_ai_grounding_metadata=[], vertex_ai_url_context_metadata=[], vertex_ai_safety_results=[], vertex_ai_citation_metadata=[], service_tier='default')"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model=None, object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='pause_turn', index=0, message=Message(content='Let me search for that information:', role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, function=Function(arguments='{\"query\": \"Solveit Answer.AI\"}', name='web_search'), id='srvtoolu_N0pThnXuYgjRcJ4HQbNLJyw8c', type='function')], function_call=None, provider_specific_fields=None), logprobs=None, provider_specific_fields=None)])"
]
},
+ "execution_count": null,
"metadata": {},
- "output_type": "display_data"
- },
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "mk_pause_web_search()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "b3a9f608",
+ "metadata": {},
+ "source": [
+ "We mock completion to return `pause_turn` in the first 2 api calls:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "95067f82",
+ "metadata": {},
+ "outputs": [
{
- "data": {
- "text/markdown": [
- "\n",
- "\n",
- "**gemini/gemini-3-flash-preview:**\n",
- "\n",
- "
"
- ],
- "text/plain": [
- "Markdown(**gemini/gemini-3-flash-preview:**)"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Mock Call 1\n",
+ "Mock Call 2\n",
+ "Mock Call 3\n",
+ "Total calls: 3\n"
+ ]
},
{
"data": {
"text/markdown": [
+ " Based on the search results, I found information about Solveit (solve.it.com), which appears to be the most prominent result. Here's a paragraph about it:\n",
"\n",
+ "Solveit is both a course and platform designed to help people solve problems using fast short iterations, covering areas like coding, writing, sysadmin, and research. The \"solveit method\" is a modern approach to building software, writing, solving problems, and learning, inspired by George Pólya's \"How to Solve It\" and developed by Jeremy Howard and team at Answer.AI. The method is founded on building in small steps with quick iterations and immediate feedback, and for coding specifically, involves writing 1-2 lines of code at a time and immediately showing the results. The platform provides users with personal instances—full virtual private servers where they can install software, store files, and host applications. It serves as an antidote to AI fatigue, helping users avoid being overwhelmed by AI-generated code they don't understand, and has been tested with over 1000 preview users.\n",
"\n",
- "🔧 simple_add({\"b\": 3, \"a\": 5})\n",
+ "🔧 web_search({\"query\": \"Solveit\"})\n",
"\n",
"\n",
"\n",
"\n",
"- id: `chatcmpl-xxx`\n",
- "- model: `gemini-3-flash-preview`\n",
- "- finish_reason: `tool_calls`\n",
- "- usage: `Usage(completion_tokens=101, prompt_tokens=93, total_tokens=194, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=83, rejected_prediction_tokens=None, text_tokens=18, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=93, image_tokens=None, video_tokens=None), cache_read_input_tokens=None)`\n",
+ "- model: `claude-sonnet-4-5-20250929`\n",
+ "- finish_reason: `stop`\n",
+ "- usage: `Usage(completion_tokens=386, prompt_tokens=12333, total_tokens=12719, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=386, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), server_tool_use=ServerToolUse(web_search_requests=1, tool_search_requests=None), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None)`\n",
"\n",
" "
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-flash-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=None, role='assistant', tool_calls=[{'index': 0, 'provider_specific_fields': {'thought_signature': 'ErECCq4CAQw51scxvuDDQXmcL7GzOo6NP4Bg2i3zLWONH5QvmgAaweGEyOeXBOo9nvqFIqf5Gr5q2q1olVKkSYpVxFumQ7YDF8nKsCOq9asYuhIKIWG+THk+CWLjwRehn0dl6+cQu8JMiZSd/+N050qcw2ADBUJEpud7ennVm+2AL5au4BDm4f1EsZDfuvrSRWXSIKF1yN0uYiLOEpmQ+Vq1wwfm/7Kawjact698X/b5fWMATY+aqT2IukpBzgxj31BZ9AvOtcUyHzOrws0J5zLASFPbh0MmAlvKPNWlqnXjBKiE5rYumemMuNwO0mwTGQAsZeWPjosWh4cMTN+iV6rsgbf2FtZFehLzxPxh+S7mWU2saocuZNzw3f08q24PzJns95M+Vneo1kxo06+VWGJYmho='}, 'function': {'arguments': '{\"b\": 3, \"a\": 5}', 'name': 'simple_add'}, 'id': 'call_6a89b537c3d84821a6090c9b35cf__thought__ErECCq4CAQw51scxvuDDQXmcL7GzOo6NP4Bg2i3zLWONH5QvmgAaweGEyOeXBOo9nvqFIqf5Gr5q2q1olVKkSYpVxFumQ7YDF8nKsCOq9asYuhIKIWG+THk+CWLjwRehn0dl6+cQu8JMiZSd/+N050qcw2ADBUJEpud7ennVm+2AL5au4BDm4f1EsZDfuvrSRWXSIKF1yN0uYiLOEpmQ+Vq1wwfm/7Kawjact698X/b5fWMATY+aqT2IukpBzgxj31BZ9AvOtcUyHzOrws0J5zLASFPbh0MmAlvKPNWlqnXjBKiE5rYumemMuNwO0mwTGQAsZeWPjosWh4cMTN+iV6rsgbf2FtZFehLzxPxh+S7mWU2saocuZNzw3f08q24PzJns95M+Vneo1kxo06+VWGJYmho=', 'type': 'function'}], function_call=None, images=[], thinking_blocks=[], provider_specific_fields={'thought_signatures': ['ErECCq4CAQw51scxvuDDQXmcL7GzOo6NP4Bg2i3zLWONH5QvmgAaweGEyOeXBOo9nvqFIqf5Gr5q2q1olVKkSYpVxFumQ7YDF8nKsCOq9asYuhIKIWG+THk+CWLjwRehn0dl6+cQu8JMiZSd/+N050qcw2ADBUJEpud7ennVm+2AL5au4BDm4f1EsZDfuvrSRWXSIKF1yN0uYiLOEpmQ+Vq1wwfm/7Kawjact698X/b5fWMATY+aqT2IukpBzgxj31BZ9AvOtcUyHzOrws0J5zLASFPbh0MmAlvKPNWlqnXjBKiE5rYumemMuNwO0mwTGQAsZeWPjosWh4cMTN+iV6rsgbf2FtZFehLzxPxh+S7mWU2saocuZNzw3f08q24PzJns95M+Vneo1kxo06+VWGJYmho=']}))], usage=Usage(completion_tokens=101, prompt_tokens=93, total_tokens=194, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=83, rejected_prediction_tokens=None, text_tokens=18, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=93, image_tokens=None, video_tokens=None), cache_read_input_tokens=None), vertex_ai_grounding_metadata=[], vertex_ai_url_context_metadata=[], vertex_ai_safety_results=[], vertex_ai_citation_metadata=[], service_tier='default')"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-5-20250929', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content=' Based on the search results, I found information about Solveit (solve.it.com), which appears to be the most prominent result. Here\\'s a paragraph about it:\\n\\nSolveit is both a course and platform designed to help people solve problems using fast short iterations, covering areas like coding, writing, sysadmin, and research. The \"solveit method\" is a modern approach to building software, writing, solving problems, and learning, inspired by George Pólya\\'s \"How to Solve It\" and developed by Jeremy Howard and team at Answer.AI. The method is founded on building in small steps with quick iterations and immediate feedback, and for coding specifically, involves writing 1-2 lines of code at a time and immediately showing the results. The platform provides users with personal instances—full virtual private servers where they can install software, store files, and host applications. It serves as an antidote to AI fatigue, helping users avoid being overwhelmed by AI-generated code they don\\'t understand, and has been tested with over 1000 preview users.', role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, function=Function(arguments='{\"query\": \"Solveit\"}', name='web_search'), id='srvtoolu_01VCLaffd3zxyZdvqNV1xkQ2', type='function')], function_call=None, provider_specific_fields={'citations': [[{'type': 'web_search_result_location', 'cited_text': 'You can now sign up for Solveit, which a course in how to solve problems (including coding, writing, sysadmin, and research) using fast short iteratio...', 'url': 'https://www.answer.ai/posts/2025-10-01-solveit-full.html', 'title': 'Launching Solveit, the antidote to AI fatigue – Answer.AI', 'encrypted_index': 'Eo8BCioIDxgCIiQ4ODk4YTFkYy0yMTNkLTRhNmYtOTljYi03ZTBlNTUzZDc0NWISDCC8srdB1wUvAaXEPRoMMKcmTvV4OJxqSmdSIjAJ0zd6pe2BkLp2sdeDWX8/x0JbDgCEa4pObOskY9bPbPM/vik9iTtCdibVIC3bxKkqE0WdsD5kpeqsl95k+iIfsvEMgfsYBA==', 'supported_text': 'Solveit is both a course and platform designed to help people solve problems using fast short iterations, covering areas like coding, writing, sysadmin, and research.'}], [{'type': 'web_search_result_location', 'cited_text': 'I want to share with you something that, for me at least, has turned out to be an effective antidote: the \"solveit method\", a modern approach to build...', 'url': 'https://solve.it.com/', 'title': 'Solve It With Code', 'encrypted_index': 'EpABCioIDxgCIiQ4ODk4YTFkYy0yMTNkLTRhNmYtOTljYi03ZTBlNTUzZDc0NWISDBmdiaXap6wP7kJwdRoMDYQz1iPEIktn2YOtIjAUanryfHdCztyK33RnWxrT28uAAEWjIJErZTAQQlFpUtmHN/HahQEeKIeNA/As8x0qFLaF22v3fK5MNCiFqyPs2sdT0oS9GAQ=', 'supported_text': 'The \"solveit method\" is a modern approach to building software, writing, solving problems, and learning, inspired by George Pólya\\'s \"How to Solve It\"'}], [{'type': 'web_search_result_location', 'cited_text': 'The Solveit method is founded in building in small steps, with quick iterations, and immediate feedback. ', 'url': 'https://solve.it.com/', 'title': 'Solve It With Code', 'encrypted_index': 'EpABCioIDxgCIiQ4ODk4YTFkYy0yMTNkLTRhNmYtOTljYi03ZTBlNTUzZDc0NWISDOAwBEDkmABorHfSghoMNbC4vObuxsQdOGZsIjAc9p50E3UkZKbpGLYQH7UOBA06WIFbdYKLvt4Qu6p5ZaaAHqaQHZekTfGsxvJYwlkqFI/Ng3F/9fNpBBs6ciOBsDm0eLWMGAQ=', 'supported_text': 'The method is founded on building in small steps with quick iterations and immediate feedback, and for coding specifically, involves writing 1-2 lines of code at a time and immediately showing the results.'}, {'type': 'web_search_result_location', 'cited_text': 'For instance, for coding, the approach is based on the system we have developed over the last decade with nbdev. We write 1-2 lines of code at a time,...', 'url': 'https://solve.it.com/', 'title': 'Solve It With Code', 'encrypted_index': 'EpMBCioIDxgCIiQ4ODk4YTFkYy0yMTNkLTRhNmYtOTljYi03ZTBlNTUzZDc0NWISDE3ReXcCBPNSTlIIVhoMK70aJSxkw26doD4zIjCq6ne7VA6qrnt+8D/aFPrTn5a/FkI53W8IgFxgsHgP0J/xc4OkdkC6VCl8woOl1bUqFxofZjMeAh3zGsNG9qrH4GiUQ1atj4vmGAQ=', 'supported_text': 'The method is founded on building in small steps with quick iterations and immediate feedback, and for coding specifically, involves writing 1-2 lines of code at a time and immediately showing the results.'}], [{'type': 'web_search_result_location', 'cited_text': 'An instance is your personal machine on which SolveIt runs. It’s a full virtual private server where you can install software, store files, and host a...', 'url': 'https://www.fast.ai/posts/2025-11-07-solveit-features.html', 'title': 'fast.ai - A Guide to Solveit Features', 'encrypted_index': 'EpEBCioIDxgCIiQ4ODk4YTFkYy0yMTNkLTRhNmYtOTljYi03ZTBlNTUzZDc0NWISDKpSJQLHkPYB7U5bahoMTyicEDCfNdGRSr2bIjBinugIV4MtfttAEqJFaw0RZiUs8JogZrWJJrRYmrJsMcPnVsnRN9SMP7SKJo2RNuMqFSWDHcLcmknK3SqHANVtfBysISEUrRgE', 'supported_text': 'The platform provides users with personal instances—full virtual private servers where they can install software, store files, and host applications.'}], [{'type': 'web_search_result_location', 'cited_text': 'It’s easier than ever to get started, but also easier than ever to let AI steer you into a situation where you’re overwhelmed by code you don’t unders...', 'url': 'https://www.answer.ai/posts/2025-10-01-solveit-full.html', 'title': 'Launching Solveit, the antidote to AI fatigue – Answer.AI', 'encrypted_index': 'EpEBCioIDxgCIiQ4ODk4YTFkYy0yMTNkLTRhNmYtOTljYi03ZTBlNTUzZDc0NWISDGgJQUovkBflUBq9tBoMPvcjHSbfagZ8o04aIjCSFbqI5pbxHTlex1P/fjFO1oyZcshLoQv0cEzV8eLzmyiu97QJ6tHdrRTQpIj0vLwqFZvp1AkoLBn6f5CQTX1cuRcxLciTohgE', 'supported_text': \"It serves as an antidote to AI fatigue, helping users avoid being overwhelmed by AI-generated code they don't understand, and has been tested with over 1000 preview users.\"}]], 'thinking_blocks': None, 'web_search_results': [{'type': 'web_search_tool_result', 'tool_use_id': 'srvtoolu_01VCLaffd3zxyZdvqNV1xkQ2', 'content': [{'type': 'web_search_result', 'title': 'Solve It With Code', 'url': 'https://solve.it.com/', 'encrypted_content': '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', 'page_age': None}, {'type': 'web_search_result', 'title': 'SolveIt | Full-Cycle Software Development Company', 'url': 'https://solveit.dev/', 'encrypted_content': 'ErsJCioIDxgCIiQ4ODk4YTFkYy0yMTNkLTRhNmYtOTljYi03ZTBlNTUzZDc0NWISDMX8G0dl7hOEtrP3TBoME5JDDiy4iMJfq2XuIjDmvPEWWqK/7o0e2/a6/sOlkRsDOV2tngjzf7ZMmdnVvuS+5V/QQ+v5C9frk6Rvr9MqvgjaHPkAvZK9IJ0YyjDUHJqdsEeliSzjGuHr7bzWKddMYeE6Tdk6xuMG+JQQWjigonA4NEizGH62Wog2cDS+JVEedm8PhHKYghGhYg7zYJ63r1VjsOhCGAA0x+PqX+r5AE8V9a88/SO4LWQtUh5J81cqr9ZfLd5ztTmSov3rmPKPENfUB6/udVobvW0zEuRuWSvF4B0sdSt3AtRuW+iCcQbD4ZhGwDaIJSoCjhs4GHfNXDQ0A1DiYXtIOjvKKP7gGdt0gGaCj/r+V49sj2Q37v75fhkJiNOE5Rxhav3wQHBJEYQOqzhMVoT0OjSdEwZDIHnBo1PKEHrxhvR92+SecnTsCOlwliYXfOi7Pg8IwoDANEupg73PYJ3KpuRyMfrlJ1EAsQQJDij91O3Jgt9HuzYBnSsDwqz0FwcsbkDZ1vNvXOGd9GurIlrMRap4WjjO6DoNX/TFQVjXUQh2rAfL7Crqb08qx7uJQJ90Y6f3rCq3lkC4GD/dgFB6We4GuFIyIidpQq1r9FMSYSO7WN4uDqQSPMiSAMrfgHhAQqHPZGuQsKKEhHOCp0rA+EWlvZ3rFv3l56fByAsOkBp7A30JkcTpQYGNLcv2ZIr6JlsSPbX3PjCfScbRTe1vD4vc2MXqLiYepPtrYL8xHt2/2vnCMaZRCpyRY5zVWHeTYB4adBvJYvLNUQvcuL7e13a/rqG3sF13w8qZNkGfErng2DvMRvldFhAKsUG7UW2JeIx/z468GyErpFzVZ+MiWGiDXzI6nX7MsBJEh4voTPSSyNUZmYvLffVgnAEZvou9ker2l4sAovWTbfg32DTUTIVxpBNCKnAyU8bPFnHVBxo5h48WHSlfXyJlMjHT/sMNMkIGZ1WTWyc3Wu2o3oD2GZNRWe8zU2m1cppHAr2tyrnUBcNvKRUK/xe1Y0I6u/+Z2667LSk2gxdVify+LmHR2LEp7u6cgcXjd0sEUDjTvwXxVhY82Tl7Fy5T9SdmImPr1aLrfqJ0Zs8bmY2LOpFmi5ELNz3cDnTKnhqN7Ka2oPfurC+sEaNhjgq66mfGKOW7faBFd72jtvu7Ni46igUt7zg7yy5GI0edkEYAo4eyWPgdjZH6va1Rn79Gi3o2qaF6ZXdamxHb2gduwr9HtWki+rwnkmSMQMRvtwLIOIw180Eg2RO1nOGv0nB3zaUiL76Q9W6MfJE4qLON3T7DUVjoSVz3YL9c4XjHJyjSgi0OuIgqIA9donufJbB1+ev3dSCLM8zJWKpCwEkNS6HMLR4xcRVd6g6q23D+457ddfFmkMuzJ+CiftT7NS/u50DIfoMf5kfOzFaNs12X1iCJ0ot7aPt261oszFqiRkZJC2DN7F3aWToxfeHKf/nT5rHxvPcH0UhI4NT/Cd5sERNxC45r//IQr1PwOwJ6Rlg6U/jCBP/T0Z1nLknAf2YyxwQTIs4o1jGSxl0YAw==', 'page_age': None}, {'type': 'web_search_result', 'title': 'fast.ai - A Guide to Solveit Features', 'url': 'https://www.fast.ai/posts/2025-11-07-solveit-features.html', 'encrypted_content': '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', 'page_age': 'November 7, 2025'}, {'type': 'web_search_result', 'title': '- Solve IT Solutions LLC', 'url': 'https://solve-it-sol.com/', 'encrypted_content': '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', 'page_age': 'April 26, 2024'}, {'type': 'web_search_result', 'title': 'Solveit – A course and platform for solving problems with code | Hacker News', 'url': 'https://news.ycombinator.com/item?id=45455719', 'encrypted_content': 'EoEGCioIDxgCIiQ4ODk4YTFkYy0yMTNkLTRhNmYtOTljYi03ZTBlNTUzZDc0NWISDNgC/UpZaVvMcxPS9RoM8Vy0ZixYaG1ElWJiIjDHUrcyNVxPS9CW6tIPIs2MCWZC6lUuSbftaiv46jr0VZ5SHFD4gS31WCjImm4Fq/EqhAUrmjWKKElZDJLFx0fIfnsN6UZkc64emcv1GqRU0gtcPPcA0fbhH5JNMDcLKT9kYlmpib7QjMp8Oguw+VPvH1Axg8e+zjPiqO1OlHnXrAaGjYkfGXovk1Jb0S0qfjootdK/16qNMOD88Tm6FTpsHayVQO9nj342dnkwOIvKuP7UVNlbJy1+Mkis3vaItrgTsHeiSMF+lpHllDB+rTzCzfYG+Ylmjo1cCJQ6jcJ67gsGbte1B8M5RzHMiN1oBEcQ7wHUXGOeutFaLyKtPo7FaYJOuar17wsiq4/ed2RIyJRz63yZwd7wHU6/jyGKKie0BYDZb0bzLRAiHtUgvzh7m7LeQw1zYL1qkK9V5nWzf2J1od6rj7Ddf74U14t6oCsiPI24UkJAYjwVRbJHQvBFXeADqOXqCZWmIygdT0u/cFooszorAIgfDd9WBB3x80UozqUT83wY1SrgBUPzCYAa0/vijmog2z00uEk/fOjWerVnXVpDic+iZCVQ6iMfJBJ5eZMa4zqtFd9XqNfKpltgelEsA8A+jhUVZiQ/b0+snKuAnr5OyYU+JpvD03BbOLheNceg8acH8qJ9+YhhjglYoi/M9HYJcb7FdDVUjinbC5galtu/7g8FdO58VAxCpSFnxebjjW8B7io4poEg1nqFDYx4EgbUIJdaWCGgA9ANRYRK2aRV9N6qA7QDcyxgPOZ4FFiWmeJtDgSGIrRx7w6r2uOS+8vazThuutnz4Zcornz7BQRKQY01Z/sRxFdnF5A0gMXCjdfA2dZPY5vbWNhJvvl0yz/IQqvNzd+x7iiXyfN/ueDx+AlQGPAiDYBmhRHTsXPHlo0K4DkyP6dHvqIxN+vZLpVS9xgD', 'page_age': 'October 15, 2025'}, {'type': 'web_search_result', 'title': 'Launching Solveit, the antidote to AI fatigue – Answer.AI', 'url': 'https://www.answer.ai/posts/2025-10-01-solveit-full.html', 'encrypted_content': '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', 'page_age': 'October 2, 2025'}, {'type': 'web_search_result', 'title': 'What Is Solveit? – johnowhitaker.dev', 'url': 'https://johnowhitaker.dev/posts/solveit.html', 'encrypted_content': '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', 'page_age': 'September 12, 2025'}, {'type': 'web_search_result', 'title': 'We have moved', 'url': 'https://solveit.fast.ai/', 'encrypted_content': 'EvcBCioIDxgCIiQ4ODk4YTFkYy0yMTNkLTRhNmYtOTljYi03ZTBlNTUzZDc0NWISDLaigGgZA/VH1wYfaxoM8qR6PAp/WyFQgRoJIjCt6gy67cBGT8sfwF2TDh+is5vdHLzpTxqJLmivxYWLP4VvAJ25/2UQoK52+0dPNKoqe2mLYeqM9Ec8KulygOWndH5tGGyW2xIZrDysi6bcs4muzIBcGcC6S232LKV6Zu+GtjTc3xIcsBc+zlF/KUs/DNOLecD8TXJgLpFQ7G5iEUbRQHDV3eWZHiS8jMWlIB+IFL/nIUKPuiIBhag74uqcBFzsPw5QAAkwXMBYmxgD', 'page_age': None}, {'type': 'web_search_result', 'title': 'SolveIT | Home', 'url': 'https://solveitservicesinc.com/', 'encrypted_content': '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', 'page_age': None}, {'type': 'web_search_result', 'title': 'SolveIT Managed Services', 'url': 'https://www.solveitms.com/', 'encrypted_content': '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', 'page_age': None}], 'caller': {'type': 'direct'}}]}))], usage=Usage(completion_tokens=386, prompt_tokens=12333, total_tokens=12719, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=386, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), server_tool_use=ServerToolUse(web_search_requests=1, tool_search_requests=None), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None))"
]
},
"metadata": {},
"output_type": "display_data"
- },
- {
- "data": {
+ }
+ ],
+ "source": [
+ "orig_completion = completion\n",
+ "\n",
+ "call_count = 0\n",
+ "def patched_completion(*args, **kwargs):\n",
+ " global call_count\n",
+ " call_count += 1\n",
+ " print(f\"Mock Call {call_count}\")\n",
+ " if call_count < 3: return mk_pause_web_search()\n",
+ " return orig_completion(*args, **kwargs)\n",
+ "\n",
+ "completion = patched_completion\n",
+ "chat_pause = Chat('claude-sonnet-4-5', search='l')\n",
+ "res = chat_pause(\"Search the web and tell me about Solveit in a paragraph\")\n",
+ "print(f\"Total calls: {call_count}\")\n",
+ "display(res)\n",
+ "\n",
+ "completion = orig_completion"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "e0062c45",
+ "metadata": {},
+ "source": [
+ "Test next turn:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "1fbb7a31",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "test_eq(len(chat_pause.hist), 4)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "fb73be0e",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Mock Call 4\n"
+ ]
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "You asked me to search the web and tell you about Solveit in a paragraph.\n",
+ "\n",
+ "\n",
+ "\n",
+ "- id: `chatcmpl-xxx`\n",
+ "- model: `claude-sonnet-4-5-20250929`\n",
+ "- finish_reason: `stop`\n",
+ "- usage: `Usage(completion_tokens=22, prompt_tokens=10334, total_tokens=10356, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=22, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None)`\n",
+ "\n",
+ " "
+ ],
"text/plain": [
- "'8'"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-5-20250929', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='You asked me to search the web and tell you about Solveit in a paragraph.', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=22, prompt_tokens=10334, total_tokens=10356, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=22, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None))"
]
},
+ "execution_count": null,
"metadata": {},
- "output_type": "display_data"
- },
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "chat_pause('What did I just ask you about?')"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "bb3f1f79",
+ "metadata": {},
+ "source": [
+ "Workaround for https://github.com/BerriAI/litellm/issues/23047:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "edc17903",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "#| export\n",
+ "import litellm.llms.anthropic.chat.transformation as _anth_t\n",
+ "import litellm.litellm_core_utils.prompt_templates.factory as _fact"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "e3278219",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "#| export\n",
+ "# if '_orig_tp' not in globals(): _orig_tp = _anth_t.AnthropicConfig.transform_parsed_response\n",
+ "# def _patched_tp(self, completion_response, raw_response, model_response, **kw):\n",
+ "# r = _orig_tp(self, completion_response, raw_response, model_response, **kw)\n",
+ "# msg = r.choices[0].message\n",
+ "# psf = getattr(msg, 'provider_specific_fields', None) or {}\n",
+ "# if len(psf.get('web_search_results') or []) > 1 and getattr(msg, 'thinking_blocks', None):\n",
+ "# psf['_original_content'] = completion_response.get('content')\n",
+ "# return r\n",
+ "# \n",
+ "# if '_orig_apt' not in globals(): _orig_apt = _fact.anthropic_messages_pt\n",
+ "# def _patched_apt(messages, model, llm_provider):\n",
+ "# for m in messages:\n",
+ "# if not hasattr(m, 'get') or m.get('role') != 'assistant': continue\n",
+ "# psf = m.get('provider_specific_fields')\n",
+ "# if not isinstance(psf, dict) or not (oc := psf.get('_original_content')): continue\n",
+ "# cc = next((b['cache_control'] for b in (m.get('content') or [])\n",
+ "# if isinstance(b, dict) and 'cache_control' in b), None)\n",
+ "# m['content'],m['thinking_blocks'] = oc,None\n",
+ "# m['tool_calls'] = [tc for tc in (m.get('tool_calls') or [])\n",
+ "# if not (getattr(tc, 'id', '') or tc.get('id', '')).startswith('srvtoolu_')] or None\n",
+ "# for k in ('web_search_results', 'tool_results'): psf.pop(k, None)\n",
+ "# if cc: oc[-1]['cache_control'] = cc\n",
+ "# return _orig_apt(messages, model, llm_provider)\n",
+ "# \n",
+ "# _anth_t.AnthropicConfig.transform_parsed_response = _patched_tp\n",
+ "# _fact.anthropic_messages_pt = _patched_apt"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "d2bdb7c7",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Turn 1: thinking=True, tcs=[ChatCompletionMessageToolCall(index=2, function=Function(arguments='{\"query\": \"fast.ai latest news 2026\"}', name='web_search'), id='srvtoolu_01JZtcwb1b8xxxBMsfDE6tBt', type='function'), ChatCompletionMessageToolCall(index=3, function=Function(arguments='{\"query\": \"answer.ai latest news 2026\"}', name='web_search'), id='srvtoolu_01RgBPcesVTnc25Yc16XibJ5', type='function')]\n",
+ "Turn 2 OK\n"
+ ]
+ }
+ ],
+ "source": [
+ "m = 'claude-sonnet-4-6'\n",
+ "msgs = [{'role':'user','content':\"Search web for latest news about fast.ai and answer.ai.\"}]\n",
+ "r = completion(m, msgs, web_search_options={\"search_context_size\":\"low\"}, reasoning_effort='low')\n",
+ "m1 = r.choices[0].message\n",
+ "print(f\"Turn 1: thinking={bool(m1.thinking_blocks)}, tcs={m1.tool_calls}\")\n",
+ "\n",
+ "msgs.append(m1)\n",
+ "msgs.append({'role':'user','content':'And search for news about solveit.'})\n",
+ "r2 = completion(m, msgs, web_search_options={\"search_context_size\":\"low\"}, reasoning_effort='low')\n",
+ "print(\"Turn 2 OK\")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "537c93a5",
+ "metadata": {},
+ "outputs": [
{
"data": {
"text/markdown": [
+ "123 + 456 = **579**. ([fast.ai](https://www.fast.ai/?utm_source=openai))\n",
"\n",
+ "I searched for **today’s fast.ai news** and didn’t find any clearly dated **Monday, April 27, 2026** news item from fast.ai itself. The fastest/most reliable signal I found was the official fast.ai homepage, which was crawled today and shows the most recent published post as **“I Don’t Want a Learning Dashboard for My Child”** from **February 17, 2026**. It also lists **“Breaking the Spell of Vibe Coding”** from **January 28, 2026** and **“How To Use AI for the Ancient Art of Close Reading”** from **January 21, 2026**. ([fast.ai](https://www.fast.ai/?utm_source=openai))\n",
"\n",
- "🔧 simple_add({\"a\": 8, \"b\": 7})\n",
+ "So the short answer is: **no obvious same-day fast.ai news showed up in web results today**; the latest visible official updates I found are those early-2026 blog posts. ([fast.ai](https://www.fast.ai/?utm_source=openai))\n",
"\n",
+ "If you want, I can do a second pass focused specifically on:\n",
+ "1. **official fast.ai/Answer.AI announcements**, \n",
+ "2. **Jeremy Howard news/interviews**, or \n",
+ "3. **GitHub/forum activity from the last 7 days**.\n",
"\n",
"\n",
"\n",
"- id: `chatcmpl-xxx`\n",
- "- model: `gemini-3-flash-preview`\n",
- "- finish_reason: `tool_calls`\n",
- "- usage: `Usage(completion_tokens=29, prompt_tokens=208, total_tokens=237, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=11, rejected_prediction_tokens=None, text_tokens=18, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=208, image_tokens=None, video_tokens=None), cache_read_input_tokens=None)`\n",
+ "- model: `gpt-5.4`\n",
+ "- finish_reason: `stop`\n",
+ "- usage: `Usage(completion_tokens=741, prompt_tokens=18567, total_tokens=19308, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=409, rejected_prediction_tokens=None, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=4608, text_tokens=None, image_tokens=None, video_tokens=None))`\n",
"\n",
" "
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-flash-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=None, role='assistant', tool_calls=[{'index': 0, 'provider_specific_fields': {'thought_signature': 'Ek4KTAEMOdbHNHzg5eUCiBSkSaMTQkvwIvn+wdWVXT9cx3+mnucgacnjpmmmC9PLAJRMoGzpXTQzMncgo5tXRJm7G3T+6OvYlpYswtD2CH0='}, 'function': {'arguments': '{\"a\": 8, \"b\": 7}', 'name': 'simple_add'}, 'id': 'call_2c09574dc6544b01b4ac8bcca27a__thought__Ek4KTAEMOdbHNHzg5eUCiBSkSaMTQkvwIvn+wdWVXT9cx3+mnucgacnjpmmmC9PLAJRMoGzpXTQzMncgo5tXRJm7G3T+6OvYlpYswtD2CH0=', 'type': 'function'}], function_call=None, images=[], thinking_blocks=[], provider_specific_fields={'thought_signatures': ['Ek4KTAEMOdbHNHzg5eUCiBSkSaMTQkvwIvn+wdWVXT9cx3+mnucgacnjpmmmC9PLAJRMoGzpXTQzMncgo5tXRJm7G3T+6OvYlpYswtD2CH0=']}))], usage=Usage(completion_tokens=29, prompt_tokens=208, total_tokens=237, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=11, rejected_prediction_tokens=None, text_tokens=18, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=208, image_tokens=None, video_tokens=None), cache_read_input_tokens=None), vertex_ai_grounding_metadata=[], vertex_ai_url_context_metadata=[], vertex_ai_safety_results=[], vertex_ai_citation_metadata=[], service_tier='default')"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gpt-5.4', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='123 + 456 = **579**. ([fast.ai](https://www.fast.ai/?utm_source=openai))\\n\\nI searched for **today’s fast.ai news** and didn’t find any clearly dated **Monday, April 27, 2026** news item from fast.ai itself. The fastest/most reliable signal I found was the official fast.ai homepage, which was crawled today and shows the most recent published post as **“I Don’t Want a Learning Dashboard for My Child”** from **February 17, 2026**. It also lists **“Breaking the Spell of Vibe Coding”** from **January 28, 2026** and **“How To Use AI for the Ancient Art of Close Reading”** from **January 21, 2026**. ([fast.ai](https://www.fast.ai/?utm_source=openai))\\n\\nSo the short answer is: **no obvious same-day fast.ai news showed up in web results today**; the latest visible official updates I found are those early-2026 blog posts. ([fast.ai](https://www.fast.ai/?utm_source=openai))\\n\\nIf you want, I can do a second pass focused specifically on:\\n1. **official fast.ai/Answer.AI announcements**, \\n2. **Jeremy Howard news/interviews**, or \\n3. **GitHub/forum activity from the last 7 days**.', role='assistant', tool_calls=None, function_call=None, reasoning_content='', reasoning_items=[{'type': 'reasoning', 'id': 'rs_01c02d6998ad23ac0069eebd47e01c819398adfb4d16dfc8ae', 'encrypted_content': None, 'summary': []}], provider_specific_fields=None, annotations=[{'type': 'url_citation', 'end_index': 72, 'start_index': 21, 'title': 'fast.ai - fast.ai—Making neural nets uncool again', 'url': 'https://www.fast.ai/?utm_source=openai'}, {'type': 'url_citation', 'end_index': 650, 'start_index': 599, 'title': 'fast.ai - fast.ai—Making neural nets uncool again', 'url': 'https://www.fast.ai/?utm_source=openai'}, {'type': 'url_citation', 'end_index': 873, 'start_index': 822, 'title': 'fast.ai - fast.ai—Making neural nets uncool again', 'url': 'https://www.fast.ai/?utm_source=openai'}]))], usage=Usage(completion_tokens=741, prompt_tokens=18567, total_tokens=19308, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=409, rejected_prediction_tokens=None, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=4608, text_tokens=None, image_tokens=None, video_tokens=None)))"
]
},
+ "execution_count": null,
"metadata": {},
- "output_type": "display_data"
- },
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "chat = Chat(gpt54, tools=[simple_add])\n",
+ "chat(\"What's 123+456? Then search the web for today's news about fast.ai.\", think='l', search='l')"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "d8cce2d9",
+ "metadata": {},
+ "source": [
+ "### Multi tool calling"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "bdb42380",
+ "metadata": {},
+ "source": [
+ "We can let the model call multiple tools in sequence using the `max_steps` parameter."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "662f7aa3",
+ "metadata": {},
+ "outputs": [
{
"data": {
+ "text/markdown": [
+ "\n",
+ "\n",
+ "**gemini/gemini-3-pro-preview:**\n",
+ "\n",
+ "
"
+ ],
"text/plain": [
- "'15'"
+ "Markdown(**gemini/gemini-3-pro-preview:**)"
]
},
"metadata": {},
@@ -6785,20 +7547,20 @@
"text/markdown": [
"\n",
"\n",
- "🔧 simple_add({\"a\": 15, \"b\": 11})\n",
+ "🔧 simple_add({\"a\": 5, \"b\": 3})\n",
"\n",
"\n",
"\n",
"\n",
"- id: `chatcmpl-xxx`\n",
- "- model: `gemini-3-flash-preview`\n",
+ "- model: `gemini-3-pro-preview`\n",
"- finish_reason: `tool_calls`\n",
- "- usage: `Usage(completion_tokens=33, prompt_tokens=252, total_tokens=285, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=13, rejected_prediction_tokens=None, text_tokens=20, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=252, image_tokens=None, video_tokens=None), cache_read_input_tokens=None)`\n",
+ "- usage: `Usage(completion_tokens=174, prompt_tokens=93, total_tokens=267, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=156, rejected_prediction_tokens=None, text_tokens=18, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=93, image_tokens=None, video_tokens=None), cache_read_input_tokens=None)`\n",
"\n",
" "
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-flash-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=None, role='assistant', tool_calls=[{'index': 0, 'provider_specific_fields': {'thought_signature': 'ElAKTgEMOdbHC2Ie/B8HxwS/HknYlx4yXymjtpX9HhjK4ON1m177KLIGhWCIIS0q7SDUTRu4UNIevu1J9t/h1UT78TGveYOscuwbZ/eo1srKfA=='}, 'function': {'arguments': '{\"a\": 15, \"b\": 11}', 'name': 'simple_add'}, 'id': 'call_15587d26a5df4965b9a4c5b5cabb__thought__ElAKTgEMOdbHC2Ie/B8HxwS/HknYlx4yXymjtpX9HhjK4ON1m177KLIGhWCIIS0q7SDUTRu4UNIevu1J9t/h1UT78TGveYOscuwbZ/eo1srKfA==', 'type': 'function'}], function_call=None, images=[], thinking_blocks=[], provider_specific_fields={'thought_signatures': ['ElAKTgEMOdbHC2Ie/B8HxwS/HknYlx4yXymjtpX9HhjK4ON1m177KLIGhWCIIS0q7SDUTRu4UNIevu1J9t/h1UT78TGveYOscuwbZ/eo1srKfA==']}))], usage=Usage(completion_tokens=33, prompt_tokens=252, total_tokens=285, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=13, rejected_prediction_tokens=None, text_tokens=20, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=252, image_tokens=None, video_tokens=None), cache_read_input_tokens=None), vertex_ai_grounding_metadata=[], vertex_ai_url_context_metadata=[], vertex_ai_safety_results=[], vertex_ai_citation_metadata=[], service_tier='default')"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-pro-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=None, role='assistant', tool_calls=[{'index': 0, 'provider_specific_fields': {'thought_signature': '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'}, 'function': {'arguments': '{\"a\": 5, \"b\": 3}', 'name': 'simple_add'}, 'id': 'call_9ca877eb965241b9a51f037da8b1__thought__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', 'type': 'function'}], function_call=None, images=[], thinking_blocks=[], provider_specific_fields={'thought_signatures': ['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']}))], usage=Usage(completion_tokens=174, prompt_tokens=93, total_tokens=267, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=156, rejected_prediction_tokens=None, text_tokens=18, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=93, image_tokens=None, video_tokens=None), cache_read_input_tokens=None), vertex_ai_grounding_metadata=[], vertex_ai_url_context_metadata=[], vertex_ai_safety_results=[], vertex_ai_citation_metadata=[])"
]
},
"metadata": {},
@@ -6807,7 +7569,7 @@
{
"data": {
"text/plain": [
- "'26'"
+ "'8'"
]
},
"metadata": {},
@@ -6816,19 +7578,22 @@
{
"data": {
"text/markdown": [
- "The result of ((5 + 3) + 7) + 11 is 26.\n",
+ "\n",
+ "\n",
+ "🔧 simple_add({\"b\": 7, \"a\": 8})\n",
+ "\n",
"\n",
"\n",
"\n",
"- id: `chatcmpl-xxx`\n",
- "- model: `gemini-3-flash-preview`\n",
- "- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=22, prompt_tokens=300, total_tokens=322, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=None, rejected_prediction_tokens=None, text_tokens=22, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=300, image_tokens=None, video_tokens=None), cache_read_input_tokens=None)`\n",
+ "- model: `gemini-3-pro-preview`\n",
+ "- finish_reason: `tool_calls`\n",
+ "- usage: `Usage(completion_tokens=42, prompt_tokens=281, total_tokens=323, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=24, rejected_prediction_tokens=None, text_tokens=18, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=281, image_tokens=None, video_tokens=None), cache_read_input_tokens=None)`\n",
"\n",
" "
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-flash-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='The result of ((5 + 3) + 7) + 11 is 26.', role='assistant', tool_calls=None, function_call=None, images=[], thinking_blocks=[], provider_specific_fields=None))], usage=Usage(completion_tokens=22, prompt_tokens=300, total_tokens=322, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=None, rejected_prediction_tokens=None, text_tokens=22, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=300, image_tokens=None, video_tokens=None), cache_read_input_tokens=None), vertex_ai_grounding_metadata=[], vertex_ai_url_context_metadata=[], vertex_ai_safety_results=[], vertex_ai_citation_metadata=[], service_tier='default')"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-pro-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=None, role='assistant', tool_calls=[{'index': 0, 'provider_specific_fields': {'thought_signature': 'EmgKZgEMOdbHRr0x5vG/DkqBSn3ZpVY87OIU84H2M5vffH769pzWIU3Vqqj+ztRPpwan7T2hk6uRpn1dsByPvlHh66uMQDRxwS1LcmmefXopD97inFuFIf6Ljq+ID2AUJQjxnup+yOoYLg=='}, 'function': {'arguments': '{\"b\": 7, \"a\": 8}', 'name': 'simple_add'}, 'id': 'call_d488198fcc2f484c86ea420a6066__thought__EmgKZgEMOdbHRr0x5vG/DkqBSn3ZpVY87OIU84H2M5vffH769pzWIU3Vqqj+ztRPpwan7T2hk6uRpn1dsByPvlHh66uMQDRxwS1LcmmefXopD97inFuFIf6Ljq+ID2AUJQjxnup+yOoYLg==', 'type': 'function'}], function_call=None, images=[], thinking_blocks=[], provider_specific_fields={'thought_signatures': ['EmgKZgEMOdbHRr0x5vG/DkqBSn3ZpVY87OIU84H2M5vffH769pzWIU3Vqqj+ztRPpwan7T2hk6uRpn1dsByPvlHh66uMQDRxwS1LcmmefXopD97inFuFIf6Ljq+ID2AUJQjxnup+yOoYLg==']}))], usage=Usage(completion_tokens=42, prompt_tokens=281, total_tokens=323, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=24, rejected_prediction_tokens=None, text_tokens=18, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=281, image_tokens=None, video_tokens=None), cache_read_input_tokens=None), vertex_ai_grounding_metadata=[], vertex_ai_url_context_metadata=[], vertex_ai_safety_results=[], vertex_ai_citation_metadata=[])"
]
},
"metadata": {},
@@ -6836,15 +7601,8 @@
},
{
"data": {
- "text/markdown": [
- "\n",
- "\n",
- "**claude-sonnet-4-6:**\n",
- "\n",
- "
"
- ],
"text/plain": [
- "Markdown(**claude-sonnet-4-6:**)"
+ "'15'"
]
},
"metadata": {},
@@ -6853,11 +7611,217 @@
{
"data": {
"text/markdown": [
- "I need to compute ((5 + 3) + 7) + 11 step by step, where each step depends on the previous result.\n",
"\n",
- "**Step 1:** Compute 5 + 3:\n",
"\n",
- "🔧 simple_add({\"a\": 5, \"b\": 3})\n",
+ "🔧 simple_add({\"a\": 15, \"b\": 11})\n",
+ "\n",
+ "\n",
+ "\n",
+ "\n",
+ "- id: `chatcmpl-xxx`\n",
+ "- model: `gemini-3-pro-preview`\n",
+ "- finish_reason: `tool_calls`\n",
+ "- usage: `Usage(completion_tokens=46, prompt_tokens=338, total_tokens=384, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=26, rejected_prediction_tokens=None, text_tokens=20, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=338, image_tokens=None, video_tokens=None), cache_read_input_tokens=None)`\n",
+ "\n",
+ " "
+ ],
+ "text/plain": [
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-pro-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=None, role='assistant', tool_calls=[{'index': 0, 'provider_specific_fields': {'thought_signature': 'EmoKaAEMOdbHDR1RCph7o/YOes1BfT+NIMvXTb8bVW1jFZ65dq1EWnui1mOA2QX5LRzNNTUzDCHMckhne3kSvdaNWCqaFRLGYggV8yTAdZufWGaWNXsxbk9YI3EbCfELckuM4mkAm4TqhNFI'}, 'function': {'arguments': '{\"a\": 15, \"b\": 11}', 'name': 'simple_add'}, 'id': 'call_a83243afd66b4041bf86091b994a__thought__EmoKaAEMOdbHDR1RCph7o/YOes1BfT+NIMvXTb8bVW1jFZ65dq1EWnui1mOA2QX5LRzNNTUzDCHMckhne3kSvdaNWCqaFRLGYggV8yTAdZufWGaWNXsxbk9YI3EbCfELckuM4mkAm4TqhNFI', 'type': 'function'}], function_call=None, images=[], thinking_blocks=[], provider_specific_fields={'thought_signatures': ['EmoKaAEMOdbHDR1RCph7o/YOes1BfT+NIMvXTb8bVW1jFZ65dq1EWnui1mOA2QX5LRzNNTUzDCHMckhne3kSvdaNWCqaFRLGYggV8yTAdZufWGaWNXsxbk9YI3EbCfELckuM4mkAm4TqhNFI']}))], usage=Usage(completion_tokens=46, prompt_tokens=338, total_tokens=384, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=26, rejected_prediction_tokens=None, text_tokens=20, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=338, image_tokens=None, video_tokens=None), cache_read_input_tokens=None), vertex_ai_grounding_metadata=[], vertex_ai_url_context_metadata=[], vertex_ai_safety_results=[], vertex_ai_citation_metadata=[])"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/plain": [
+ "'26'"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "The result of ((5 + 3) + 7) + 11 is 26.\n",
+ "\n",
+ "\n",
+ "\n",
+ "- id: `chatcmpl-xxx`\n",
+ "- model: `gemini-3-pro-preview`\n",
+ "- finish_reason: `stop`\n",
+ "- usage: `Usage(completion_tokens=46, prompt_tokens=399, total_tokens=445, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=24, rejected_prediction_tokens=None, text_tokens=22, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=399, image_tokens=None, video_tokens=None), cache_read_input_tokens=None)`\n",
+ "\n",
+ " "
+ ],
+ "text/plain": [
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-pro-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='The result of ((5 + 3) + 7) + 11 is 26.', role='assistant', tool_calls=None, function_call=None, images=[], thinking_blocks=[], provider_specific_fields={'thought_signatures': ['EokBCoYBAQw51sdzis/bOYhfim+7lao+teR/v50sqVFzqb7DLCt2reNskgMZBvD7xwjS4aDD+Fumrb5uL4Sy233VRWuW7Bz4XlNXbuhwv89fl2wh3xIk42pW3lPOJN4HdbeG6zBcXtePy7Ys8sgUFLdsakr4Tvoemvn4COKUy1d09eKP5nuBRAp0m48=']}))], usage=Usage(completion_tokens=46, prompt_tokens=399, total_tokens=445, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=24, rejected_prediction_tokens=None, text_tokens=22, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=399, image_tokens=None, video_tokens=None), cache_read_input_tokens=None), vertex_ai_grounding_metadata=[], vertex_ai_url_context_metadata=[], vertex_ai_safety_results=[], vertex_ai_citation_metadata=[])"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "\n",
+ "\n",
+ "**gemini/gemini-3-flash-preview:**\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(**gemini/gemini-3-flash-preview:**)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "\n",
+ "\n",
+ "🔧 simple_add({\"b\": 3, \"a\": 5})\n",
+ "\n",
+ "\n",
+ "\n",
+ "\n",
+ "- id: `chatcmpl-xxx`\n",
+ "- model: `gemini-3-flash-preview`\n",
+ "- finish_reason: `tool_calls`\n",
+ "- usage: `Usage(completion_tokens=101, prompt_tokens=93, total_tokens=194, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=83, rejected_prediction_tokens=None, text_tokens=18, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=93, image_tokens=None, video_tokens=None), cache_read_input_tokens=None)`\n",
+ "\n",
+ " "
+ ],
+ "text/plain": [
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-flash-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=None, role='assistant', tool_calls=[{'index': 0, 'provider_specific_fields': {'thought_signature': 'ErECCq4CAQw51scxvuDDQXmcL7GzOo6NP4Bg2i3zLWONH5QvmgAaweGEyOeXBOo9nvqFIqf5Gr5q2q1olVKkSYpVxFumQ7YDF8nKsCOq9asYuhIKIWG+THk+CWLjwRehn0dl6+cQu8JMiZSd/+N050qcw2ADBUJEpud7ennVm+2AL5au4BDm4f1EsZDfuvrSRWXSIKF1yN0uYiLOEpmQ+Vq1wwfm/7Kawjact698X/b5fWMATY+aqT2IukpBzgxj31BZ9AvOtcUyHzOrws0J5zLASFPbh0MmAlvKPNWlqnXjBKiE5rYumemMuNwO0mwTGQAsZeWPjosWh4cMTN+iV6rsgbf2FtZFehLzxPxh+S7mWU2saocuZNzw3f08q24PzJns95M+Vneo1kxo06+VWGJYmho='}, 'function': {'arguments': '{\"b\": 3, \"a\": 5}', 'name': 'simple_add'}, 'id': 'call_be006d2b3f344313bac2efa79b4f__thought__ErECCq4CAQw51scxvuDDQXmcL7GzOo6NP4Bg2i3zLWONH5QvmgAaweGEyOeXBOo9nvqFIqf5Gr5q2q1olVKkSYpVxFumQ7YDF8nKsCOq9asYuhIKIWG+THk+CWLjwRehn0dl6+cQu8JMiZSd/+N050qcw2ADBUJEpud7ennVm+2AL5au4BDm4f1EsZDfuvrSRWXSIKF1yN0uYiLOEpmQ+Vq1wwfm/7Kawjact698X/b5fWMATY+aqT2IukpBzgxj31BZ9AvOtcUyHzOrws0J5zLASFPbh0MmAlvKPNWlqnXjBKiE5rYumemMuNwO0mwTGQAsZeWPjosWh4cMTN+iV6rsgbf2FtZFehLzxPxh+S7mWU2saocuZNzw3f08q24PzJns95M+Vneo1kxo06+VWGJYmho=', 'type': 'function'}], function_call=None, images=[], thinking_blocks=[], provider_specific_fields={'thought_signatures': ['ErECCq4CAQw51scxvuDDQXmcL7GzOo6NP4Bg2i3zLWONH5QvmgAaweGEyOeXBOo9nvqFIqf5Gr5q2q1olVKkSYpVxFumQ7YDF8nKsCOq9asYuhIKIWG+THk+CWLjwRehn0dl6+cQu8JMiZSd/+N050qcw2ADBUJEpud7ennVm+2AL5au4BDm4f1EsZDfuvrSRWXSIKF1yN0uYiLOEpmQ+Vq1wwfm/7Kawjact698X/b5fWMATY+aqT2IukpBzgxj31BZ9AvOtcUyHzOrws0J5zLASFPbh0MmAlvKPNWlqnXjBKiE5rYumemMuNwO0mwTGQAsZeWPjosWh4cMTN+iV6rsgbf2FtZFehLzxPxh+S7mWU2saocuZNzw3f08q24PzJns95M+Vneo1kxo06+VWGJYmho=']}))], usage=Usage(completion_tokens=101, prompt_tokens=93, total_tokens=194, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=83, rejected_prediction_tokens=None, text_tokens=18, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=93, image_tokens=None, video_tokens=None), cache_read_input_tokens=None), vertex_ai_grounding_metadata=[], vertex_ai_url_context_metadata=[], vertex_ai_safety_results=[], vertex_ai_citation_metadata=[])"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/plain": [
+ "'8'"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "\n",
+ "\n",
+ "🔧 simple_add({\"a\": 8, \"b\": 7})\n",
+ "\n",
+ "\n",
+ "\n",
+ "\n",
+ "- id: `chatcmpl-xxx`\n",
+ "- model: `gemini-3-flash-preview`\n",
+ "- finish_reason: `tool_calls`\n",
+ "- usage: `Usage(completion_tokens=29, prompt_tokens=208, total_tokens=237, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=11, rejected_prediction_tokens=None, text_tokens=18, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=208, image_tokens=None, video_tokens=None), cache_read_input_tokens=None)`\n",
+ "\n",
+ " "
+ ],
+ "text/plain": [
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-flash-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=None, role='assistant', tool_calls=[{'index': 0, 'provider_specific_fields': {'thought_signature': 'Ek4KTAEMOdbHNHzg5eUCiBSkSaMTQkvwIvn+wdWVXT9cx3+mnucgacnjpmmmC9PLAJRMoGzpXTQzMncgo5tXRJm7G3T+6OvYlpYswtD2CH0='}, 'function': {'arguments': '{\"a\": 8, \"b\": 7}', 'name': 'simple_add'}, 'id': 'call_2c0eced0429441caab93ff10a4a1__thought__Ek4KTAEMOdbHNHzg5eUCiBSkSaMTQkvwIvn+wdWVXT9cx3+mnucgacnjpmmmC9PLAJRMoGzpXTQzMncgo5tXRJm7G3T+6OvYlpYswtD2CH0=', 'type': 'function'}], function_call=None, images=[], thinking_blocks=[], provider_specific_fields={'thought_signatures': ['Ek4KTAEMOdbHNHzg5eUCiBSkSaMTQkvwIvn+wdWVXT9cx3+mnucgacnjpmmmC9PLAJRMoGzpXTQzMncgo5tXRJm7G3T+6OvYlpYswtD2CH0=']}))], usage=Usage(completion_tokens=29, prompt_tokens=208, total_tokens=237, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=11, rejected_prediction_tokens=None, text_tokens=18, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=208, image_tokens=None, video_tokens=None), cache_read_input_tokens=None), vertex_ai_grounding_metadata=[], vertex_ai_url_context_metadata=[], vertex_ai_safety_results=[], vertex_ai_citation_metadata=[])"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/plain": [
+ "'15'"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "\n",
+ "\n",
+ "🔧 simple_add({\"a\": 15, \"b\": 11})\n",
+ "\n",
+ "\n",
+ "\n",
+ "\n",
+ "- id: `chatcmpl-xxx`\n",
+ "- model: `gemini-3-flash-preview`\n",
+ "- finish_reason: `tool_calls`\n",
+ "- usage: `Usage(completion_tokens=33, prompt_tokens=252, total_tokens=285, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=13, rejected_prediction_tokens=None, text_tokens=20, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=252, image_tokens=None, video_tokens=None), cache_read_input_tokens=None)`\n",
+ "\n",
+ " "
+ ],
+ "text/plain": [
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-flash-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=None, role='assistant', tool_calls=[{'index': 0, 'provider_specific_fields': {'thought_signature': 'ElAKTgEMOdbHC2Ie/B8HxwS/HknYlx4yXymjtpX9HhjK4ON1m177KLIGhWCIIS0q7SDUTRu4UNIevu1J9t/h1UT78TGveYOscuwbZ/eo1srKfA=='}, 'function': {'arguments': '{\"a\": 15, \"b\": 11}', 'name': 'simple_add'}, 'id': 'call_69f9ec03de8a483e8e8d622fbba9__thought__ElAKTgEMOdbHC2Ie/B8HxwS/HknYlx4yXymjtpX9HhjK4ON1m177KLIGhWCIIS0q7SDUTRu4UNIevu1J9t/h1UT78TGveYOscuwbZ/eo1srKfA==', 'type': 'function'}], function_call=None, images=[], thinking_blocks=[], provider_specific_fields={'thought_signatures': ['ElAKTgEMOdbHC2Ie/B8HxwS/HknYlx4yXymjtpX9HhjK4ON1m177KLIGhWCIIS0q7SDUTRu4UNIevu1J9t/h1UT78TGveYOscuwbZ/eo1srKfA==']}))], usage=Usage(completion_tokens=33, prompt_tokens=252, total_tokens=285, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=13, rejected_prediction_tokens=None, text_tokens=20, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=252, image_tokens=None, video_tokens=None), cache_read_input_tokens=None), vertex_ai_grounding_metadata=[], vertex_ai_url_context_metadata=[], vertex_ai_safety_results=[], vertex_ai_citation_metadata=[])"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/plain": [
+ "'26'"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "The result of ((5 + 3) + 7) + 11 is 26.\n",
+ "\n",
+ "\n",
+ "\n",
+ "- id: `chatcmpl-xxx`\n",
+ "- model: `gemini-3-flash-preview`\n",
+ "- finish_reason: `stop`\n",
+ "- usage: `Usage(completion_tokens=22, prompt_tokens=300, total_tokens=322, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=None, rejected_prediction_tokens=None, text_tokens=22, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=300, image_tokens=None, video_tokens=None), cache_read_input_tokens=None)`\n",
+ "\n",
+ " "
+ ],
+ "text/plain": [
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-flash-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='The result of ((5 + 3) + 7) + 11 is 26.', role='assistant', tool_calls=None, function_call=None, images=[], thinking_blocks=[], provider_specific_fields=None))], usage=Usage(completion_tokens=22, prompt_tokens=300, total_tokens=322, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=None, rejected_prediction_tokens=None, text_tokens=22, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=300, image_tokens=None, video_tokens=None), cache_read_input_tokens=None), vertex_ai_grounding_metadata=[], vertex_ai_url_context_metadata=[], vertex_ai_safety_results=[], vertex_ai_citation_metadata=[])"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "\n",
+ "\n",
+ "**claude-sonnet-4-6:**\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(**claude-sonnet-4-6:**)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "I need to compute ((5 + 3) + 7) + 11 step by step, where each step depends on the previous result.\n",
+ "\n",
+ "**Step 1:** Compute 5 + 3:\n",
+ "\n",
+ "🔧 simple_add({\"a\": 5, \"b\": 3})\n",
"\n",
"\n",
"\n",
@@ -6865,12 +7829,12 @@
"- id: `chatcmpl-xxx`\n",
"- model: `claude-sonnet-4-6`\n",
"- finish_reason: `tool_calls`\n",
- "- usage: `Usage(completion_tokens=118, prompt_tokens=617, total_tokens=735, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=118, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=617, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
+ "- usage: `Usage(completion_tokens=118, prompt_tokens=617, total_tokens=735, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=118, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
"\n",
" "
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content='I need to compute ((5 + 3) + 7) + 11 step by step, where each step depends on the previous result.\\n\\n**Step 1:** Compute 5 + 3:', role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, caller={'type': 'direct'}, function=Function(arguments='{\"a\": 5, \"b\": 3}', name='simple_add'), id='toolu_01PBnDM1smEdjGAqu6vtHE67', type='function')], function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=118, prompt_tokens=617, total_tokens=735, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=118, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=617, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content='I need to compute ((5 + 3) + 7) + 11 step by step, where each step depends on the previous result.\\n\\n**Step 1:** Compute 5 + 3:', role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, caller={'type': 'direct'}, function=Function(arguments='{\"a\": 5, \"b\": 3}', name='simple_add'), id='toolu_01PBnDM1smEdjGAqu6vtHE67', type='function')], function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=118, prompt_tokens=617, total_tokens=735, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=118, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
]
},
"metadata": {},
@@ -6898,12 +7862,12 @@
"- id: `chatcmpl-xxx`\n",
"- model: `claude-sonnet-4-6`\n",
"- finish_reason: `tool_calls`\n",
- "- usage: `Usage(completion_tokens=90, prompt_tokens=748, total_tokens=838, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=90, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=748, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
+ "- usage: `Usage(completion_tokens=90, prompt_tokens=748, total_tokens=838, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=90, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
"\n",
" "
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content='**Step 2:** Add 7 to the result (8 + 7):', role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, caller={'type': 'direct'}, function=Function(arguments='{\"a\": 8, \"b\": 7}', name='simple_add'), id='toolu_01Tm3Yskxi31YRD4yuv33R5d', type='function')], function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=90, prompt_tokens=748, total_tokens=838, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=90, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=748, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content='**Step 2:** Add 7 to the result (8 + 7):', role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, caller={'type': 'direct'}, function=Function(arguments='{\"a\": 8, \"b\": 7}', name='simple_add'), id='toolu_01Tm3Yskxi31YRD4yuv33R5d', type='function')], function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=90, prompt_tokens=748, total_tokens=838, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=90, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
]
},
"metadata": {},
@@ -6931,12 +7895,12 @@
"- id: `chatcmpl-xxx`\n",
"- model: `claude-sonnet-4-6`\n",
"- finish_reason: `tool_calls`\n",
- "- usage: `Usage(completion_tokens=90, prompt_tokens=851, total_tokens=941, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=90, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=851, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
+ "- usage: `Usage(completion_tokens=90, prompt_tokens=851, total_tokens=941, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=90, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
"\n",
""
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content='**Step 3:** Add 11 to the result (15 + 11):', role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, caller={'type': 'direct'}, function=Function(arguments='{\"a\": 15, \"b\": 11}', name='simple_add'), id='toolu_01EaXKKMBYqo83nRTbqyKc4P', type='function')], function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=90, prompt_tokens=851, total_tokens=941, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=90, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=851, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content='**Step 3:** Add 11 to the result (15 + 11):', role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, caller={'type': 'direct'}, function=Function(arguments='{\"a\": 15, \"b\": 11}', name='simple_add'), id='toolu_01EaXKKMBYqo83nRTbqyKc4P', type='function')], function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=90, prompt_tokens=851, total_tokens=941, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=90, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
]
},
"metadata": {},
@@ -6966,12 +7930,12 @@
"- id: `chatcmpl-xxx`\n",
"- model: `claude-sonnet-4-6`\n",
"- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=74, prompt_tokens=954, total_tokens=1028, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=74, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=954, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
+ "- usage: `Usage(completion_tokens=74, prompt_tokens=954, total_tokens=1028, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=74, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
"\n",
""
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content=\"Here's the breakdown:\\n1. **5 + 3 = 8**\\n2. **8 + 7 = 15**\\n3. **15 + 11 = 26**\\n\\nTherefore, **((5 + 3) + 7) + 11 = 26**!\", role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=74, prompt_tokens=954, total_tokens=1028, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=74, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=954, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content=\"Here's the breakdown:\\n1. **5 + 3 = 8**\\n2. **8 + 7 = 15**\\n3. **15 + 11 = 26**\\n\\nTherefore, **((5 + 3) + 7) + 11 = 26**!\", role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=74, prompt_tokens=954, total_tokens=1028, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=74, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
]
},
"metadata": {},
@@ -7154,7 +8118,7 @@
""
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-flash-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=None, role='assistant', tool_calls=[{'index': 0, 'provider_specific_fields': {'thought_signature': 'EukCCuYCAQw51sdxwHpd5drdFftmAdpAMnl4hy5jlhzys8Mjy1DJTHzy9oE9u7sn4mW8HFAiDeMgSqb9JiR9awrWiuGCl1GTtUVVMHbFMM8rvlJBzVj1OV/7iSl23WxEOIXC/sdfnGgtrb77/A6ms8lCUb6VVF9yPtaQacHlGgRvm2Mwq24dgCFz/qOOcW86ytV6Gjer9gFCSLsTink45IgGkg9HcrmrMXnOvAsV0PMKnQkVxOgIspXhfeqV+B8KckgpLaBvO9qJ3TIbiHxl0mQYL5KhpS5uBeZE3JCSJVe6TqESnr9x/LUHf8FNZ+1lg2GK6uhynIh4+jd2zIIWvPmkuGbndVnRkbpbfTxS98JezIdLc/K8IRKeSt5j4arC0IBaEAjTOKNvOqDDGeSnLf9D19wAh4LVejAwsaYpTh2HhODXIUEY/b6oJUt7yzK/vtj8oR7c++eRXW2pc2IU1ZOwaX9V019OHsuqiQ=='}, 'function': {'arguments': '{\"b\": 3, \"a\": 5}', 'name': 'simple_add'}, 'id': 'call_d71dbdea4adb4e809fb0a02f7b2d__thought__EukCCuYCAQw51sdxwHpd5drdFftmAdpAMnl4hy5jlhzys8Mjy1DJTHzy9oE9u7sn4mW8HFAiDeMgSqb9JiR9awrWiuGCl1GTtUVVMHbFMM8rvlJBzVj1OV/7iSl23WxEOIXC/sdfnGgtrb77/A6ms8lCUb6VVF9yPtaQacHlGgRvm2Mwq24dgCFz/qOOcW86ytV6Gjer9gFCSLsTink45IgGkg9HcrmrMXnOvAsV0PMKnQkVxOgIspXhfeqV+B8KckgpLaBvO9qJ3TIbiHxl0mQYL5KhpS5uBeZE3JCSJVe6TqESnr9x/LUHf8FNZ+1lg2GK6uhynIh4+jd2zIIWvPmkuGbndVnRkbpbfTxS98JezIdLc/K8IRKeSt5j4arC0IBaEAjTOKNvOqDDGeSnLf9D19wAh4LVejAwsaYpTh2HhODXIUEY/b6oJUt7yzK/vtj8oR7c++eRXW2pc2IU1ZOwaX9V019OHsuqiQ==', 'type': 'function'}, {'index': 1, 'function': {'arguments': '{\"b\": 2, \"a\": 7}', 'name': 'simple_add'}, 'id': 'call_177d235506784d06b66c8885dce3', 'type': 'function'}], function_call=None, images=[], thinking_blocks=[], provider_specific_fields={'thought_signatures': ['EukCCuYCAQw51sdxwHpd5drdFftmAdpAMnl4hy5jlhzys8Mjy1DJTHzy9oE9u7sn4mW8HFAiDeMgSqb9JiR9awrWiuGCl1GTtUVVMHbFMM8rvlJBzVj1OV/7iSl23WxEOIXC/sdfnGgtrb77/A6ms8lCUb6VVF9yPtaQacHlGgRvm2Mwq24dgCFz/qOOcW86ytV6Gjer9gFCSLsTink45IgGkg9HcrmrMXnOvAsV0PMKnQkVxOgIspXhfeqV+B8KckgpLaBvO9qJ3TIbiHxl0mQYL5KhpS5uBeZE3JCSJVe6TqESnr9x/LUHf8FNZ+1lg2GK6uhynIh4+jd2zIIWvPmkuGbndVnRkbpbfTxS98JezIdLc/K8IRKeSt5j4arC0IBaEAjTOKNvOqDDGeSnLf9D19wAh4LVejAwsaYpTh2HhODXIUEY/b6oJUt7yzK/vtj8oR7c++eRXW2pc2IU1ZOwaX9V019OHsuqiQ==']}))], usage=Usage(completion_tokens=140, prompt_tokens=148, total_tokens=288, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=104, rejected_prediction_tokens=None, text_tokens=36, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=148, image_tokens=None, video_tokens=None), cache_read_input_tokens=None), vertex_ai_grounding_metadata=[], vertex_ai_url_context_metadata=[], vertex_ai_safety_results=[], vertex_ai_citation_metadata=[])"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-flash-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=None, role='assistant', tool_calls=[{'index': 0, 'provider_specific_fields': {'thought_signature': 'EukCCuYCAQw51sdxwHpd5drdFftmAdpAMnl4hy5jlhzys8Mjy1DJTHzy9oE9u7sn4mW8HFAiDeMgSqb9JiR9awrWiuGCl1GTtUVVMHbFMM8rvlJBzVj1OV/7iSl23WxEOIXC/sdfnGgtrb77/A6ms8lCUb6VVF9yPtaQacHlGgRvm2Mwq24dgCFz/qOOcW86ytV6Gjer9gFCSLsTink45IgGkg9HcrmrMXnOvAsV0PMKnQkVxOgIspXhfeqV+B8KckgpLaBvO9qJ3TIbiHxl0mQYL5KhpS5uBeZE3JCSJVe6TqESnr9x/LUHf8FNZ+1lg2GK6uhynIh4+jd2zIIWvPmkuGbndVnRkbpbfTxS98JezIdLc/K8IRKeSt5j4arC0IBaEAjTOKNvOqDDGeSnLf9D19wAh4LVejAwsaYpTh2HhODXIUEY/b6oJUt7yzK/vtj8oR7c++eRXW2pc2IU1ZOwaX9V019OHsuqiQ=='}, 'function': {'arguments': '{\"b\": 3, \"a\": 5}', 'name': 'simple_add'}, 'id': 'call_e536c624d9f445a59236c6a91e08__thought__EukCCuYCAQw51sdxwHpd5drdFftmAdpAMnl4hy5jlhzys8Mjy1DJTHzy9oE9u7sn4mW8HFAiDeMgSqb9JiR9awrWiuGCl1GTtUVVMHbFMM8rvlJBzVj1OV/7iSl23WxEOIXC/sdfnGgtrb77/A6ms8lCUb6VVF9yPtaQacHlGgRvm2Mwq24dgCFz/qOOcW86ytV6Gjer9gFCSLsTink45IgGkg9HcrmrMXnOvAsV0PMKnQkVxOgIspXhfeqV+B8KckgpLaBvO9qJ3TIbiHxl0mQYL5KhpS5uBeZE3JCSJVe6TqESnr9x/LUHf8FNZ+1lg2GK6uhynIh4+jd2zIIWvPmkuGbndVnRkbpbfTxS98JezIdLc/K8IRKeSt5j4arC0IBaEAjTOKNvOqDDGeSnLf9D19wAh4LVejAwsaYpTh2HhODXIUEY/b6oJUt7yzK/vtj8oR7c++eRXW2pc2IU1ZOwaX9V019OHsuqiQ==', 'type': 'function'}, {'index': 1, 'function': {'arguments': '{\"b\": 2, \"a\": 7}', 'name': 'simple_add'}, 'id': 'call_c13350665522409b89d03ce0fc44', 'type': 'function'}], function_call=None, images=[], thinking_blocks=[], provider_specific_fields={'thought_signatures': ['EukCCuYCAQw51sdxwHpd5drdFftmAdpAMnl4hy5jlhzys8Mjy1DJTHzy9oE9u7sn4mW8HFAiDeMgSqb9JiR9awrWiuGCl1GTtUVVMHbFMM8rvlJBzVj1OV/7iSl23WxEOIXC/sdfnGgtrb77/A6ms8lCUb6VVF9yPtaQacHlGgRvm2Mwq24dgCFz/qOOcW86ytV6Gjer9gFCSLsTink45IgGkg9HcrmrMXnOvAsV0PMKnQkVxOgIspXhfeqV+B8KckgpLaBvO9qJ3TIbiHxl0mQYL5KhpS5uBeZE3JCSJVe6TqESnr9x/LUHf8FNZ+1lg2GK6uhynIh4+jd2zIIWvPmkuGbndVnRkbpbfTxS98JezIdLc/K8IRKeSt5j4arC0IBaEAjTOKNvOqDDGeSnLf9D19wAh4LVejAwsaYpTh2HhODXIUEY/b6oJUt7yzK/vtj8oR7c++eRXW2pc2IU1ZOwaX9V019OHsuqiQ==']}))], usage=Usage(completion_tokens=140, prompt_tokens=148, total_tokens=288, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=104, rejected_prediction_tokens=None, text_tokens=36, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=148, image_tokens=None, video_tokens=None), cache_read_input_tokens=None), vertex_ai_grounding_metadata=[], vertex_ai_url_context_metadata=[], vertex_ai_safety_results=[], vertex_ai_citation_metadata=[])"
]
},
"metadata": {},
@@ -7163,7 +8127,7 @@
{
"data": {
"text/plain": [
- "{'tool_call_id': 'call_d71dbdea4adb4e809fb0a02f7b2d__thought__EukCCuYCAQw51sdxwHpd5drdFftmAdpAMnl4hy5jlhzys8Mjy1DJTHzy',\n",
+ "{'tool_call_id': 'call_e536c624d9f445a59236c6a91e08__thought__EukCCuYCAQw51sdxwHpd5drdFftmAdpAMnl4hy5jlhzys8Mjy1DJTHzy',\n",
" 'role': 'tool',\n",
" 'name': 'simple_add',\n",
" 'content': '8'}"
@@ -7175,7 +8139,7 @@
{
"data": {
"text/plain": [
- "{'tool_call_id': 'call_177d235506784d06b66c8885dce3',\n",
+ "{'tool_call_id': 'call_c13350665522409b89d03ce0fc44',\n",
" 'role': 'tool',\n",
" 'name': 'simple_add',\n",
" 'content': '9'}"
@@ -7202,7 +8166,7 @@
""
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-flash-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=None, role='assistant', tool_calls=[{'index': 0, 'provider_specific_fields': {'thought_signature': 'En4KfAEMOdbH9GJeYXSOTySH6fpIeHcprIOzRJgzxcouFAanv2M9uzOmCvYzlxJ6Y4dEY9FZqoGUSYkqt+Q33WAPMQODlOMWBMIN8KOxD3KklglXxTfndX00rD2+WllIZgRU04Rpr6XkC2d7bu4/HU56YKll9FtTNc3qx8OXwEk='}, 'function': {'arguments': '{\"a\": 8, \"b\": 9}', 'name': 'multiply'}, 'id': 'call_96ebb1ddb733450a9bc557a495bf__thought__En4KfAEMOdbH9GJeYXSOTySH6fpIeHcprIOzRJgzxcouFAanv2M9uzOmCvYzlxJ6Y4dEY9FZqoGUSYkqt+Q33WAPMQODlOMWBMIN8KOxD3KklglXxTfndX00rD2+WllIZgRU04Rpr6XkC2d7bu4/HU56YKll9FtTNc3qx8OXwEk=', 'type': 'function'}], function_call=None, images=[], thinking_blocks=[], provider_specific_fields={'thought_signatures': ['En4KfAEMOdbH9GJeYXSOTySH6fpIeHcprIOzRJgzxcouFAanv2M9uzOmCvYzlxJ6Y4dEY9FZqoGUSYkqt+Q33WAPMQODlOMWBMIN8KOxD3KklglXxTfndX00rD2+WllIZgRU04Rpr6XkC2d7bu4/HU56YKll9FtTNc3qx8OXwEk=']}))], usage=Usage(completion_tokens=37, prompt_tokens=314, total_tokens=351, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=21, rejected_prediction_tokens=None, text_tokens=16, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=314, image_tokens=None, video_tokens=None), cache_read_input_tokens=None), vertex_ai_grounding_metadata=[], vertex_ai_url_context_metadata=[], vertex_ai_safety_results=[], vertex_ai_citation_metadata=[])"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-flash-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=None, role='assistant', tool_calls=[{'index': 0, 'provider_specific_fields': {'thought_signature': 'En4KfAEMOdbH9GJeYXSOTySH6fpIeHcprIOzRJgzxcouFAanv2M9uzOmCvYzlxJ6Y4dEY9FZqoGUSYkqt+Q33WAPMQODlOMWBMIN8KOxD3KklglXxTfndX00rD2+WllIZgRU04Rpr6XkC2d7bu4/HU56YKll9FtTNc3qx8OXwEk='}, 'function': {'arguments': '{\"a\": 8, \"b\": 9}', 'name': 'multiply'}, 'id': 'call_c6128a08b0d44c2eafba56b3728d__thought__En4KfAEMOdbH9GJeYXSOTySH6fpIeHcprIOzRJgzxcouFAanv2M9uzOmCvYzlxJ6Y4dEY9FZqoGUSYkqt+Q33WAPMQODlOMWBMIN8KOxD3KklglXxTfndX00rD2+WllIZgRU04Rpr6XkC2d7bu4/HU56YKll9FtTNc3qx8OXwEk=', 'type': 'function'}], function_call=None, images=[], thinking_blocks=[], provider_specific_fields={'thought_signatures': ['En4KfAEMOdbH9GJeYXSOTySH6fpIeHcprIOzRJgzxcouFAanv2M9uzOmCvYzlxJ6Y4dEY9FZqoGUSYkqt+Q33WAPMQODlOMWBMIN8KOxD3KklglXxTfndX00rD2+WllIZgRU04Rpr6XkC2d7bu4/HU56YKll9FtTNc3qx8OXwEk=']}))], usage=Usage(completion_tokens=37, prompt_tokens=314, total_tokens=351, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=21, rejected_prediction_tokens=None, text_tokens=16, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=314, image_tokens=None, video_tokens=None), cache_read_input_tokens=None), vertex_ai_grounding_metadata=[], vertex_ai_url_context_metadata=[], vertex_ai_safety_results=[], vertex_ai_citation_metadata=[])"
]
},
"metadata": {},
@@ -7211,7 +8175,7 @@
{
"data": {
"text/plain": [
- "{'tool_call_id': 'call_96ebb1ddb733450a9bc557a495bf__thought__En4KfAEMOdbH9GJeYXSOTySH6fpIeHcprIOzRJgzxcouFAanv2M9uzOm',\n",
+ "{'tool_call_id': 'call_c6128a08b0d44c2eafba56b3728d__thought__En4KfAEMOdbH9GJeYXSOTySH6fpIeHcprIOzRJgzxcouFAanv2M9uzOm',\n",
" 'role': 'tool',\n",
" 'name': 'multiply',\n",
" 'content': '72'}"
@@ -7276,12 +8240,12 @@
"- id: `chatcmpl-xxx`\n",
"- model: `claude-sonnet-4-6`\n",
"- finish_reason: `tool_calls`\n",
- "- usage: `Usage(completion_tokens=173, prompt_tokens=701, total_tokens=874, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=173, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=701, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
+ "- usage: `Usage(completion_tokens=173, prompt_tokens=701, total_tokens=874, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=173, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
"\n",
""
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=\"I need to calculate **(5 + 3) \\\\* (7 + 2)**. I'll start by performing both additions simultaneously, then multiply the results.\\n\\n**Step 1: Compute both additions in parallel.**\", role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, caller={'type': 'direct'}, function=Function(arguments='{\"a\": 5, \"b\": 3}', name='simple_add'), id='toolu_01NhhRVainrzD11Jc356s8sE', type='function'), ChatCompletionMessageToolCall(index=2, caller={'type': 'direct'}, function=Function(arguments='{\"a\": 7, \"b\": 2}', name='simple_add'), id='toolu_01Bnir83S53xzFTRftmrr3J6', type='function')], function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=173, prompt_tokens=701, total_tokens=874, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=173, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=701, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=\"I need to calculate **(5 + 3) \\\\* (7 + 2)**. I'll start by performing both additions simultaneously, then multiply the results.\\n\\n**Step 1: Compute both additions in parallel.**\", role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, caller={'type': 'direct'}, function=Function(arguments='{\"a\": 5, \"b\": 3}', name='simple_add'), id='toolu_01NhhRVainrzD11Jc356s8sE', type='function'), ChatCompletionMessageToolCall(index=2, caller={'type': 'direct'}, function=Function(arguments='{\"a\": 7, \"b\": 2}', name='simple_add'), id='toolu_01Bnir83S53xzFTRftmrr3J6', type='function')], function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=173, prompt_tokens=701, total_tokens=874, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=173, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
]
},
"metadata": {},
@@ -7326,12 +8290,12 @@
"- id: `chatcmpl-xxx`\n",
"- model: `claude-sonnet-4-6`\n",
"- finish_reason: `tool_calls`\n",
- "- usage: `Usage(completion_tokens=111, prompt_tokens=939, total_tokens=1050, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=111, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=939, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
+ "- usage: `Usage(completion_tokens=111, prompt_tokens=939, total_tokens=1050, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=111, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
"\n",
""
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=\"5 + 3 = **8** and 7 + 2 = **9**. Now I'll multiply the two results.\\n\\n**Step 2: Multiply the results.**\", role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, caller={'type': 'direct'}, function=Function(arguments='{\"a\": 8, \"b\": 9}', name='multiply'), id='toolu_011dP6xwp1VHWRMYRLnHQH2v', type='function')], function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=111, prompt_tokens=939, total_tokens=1050, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=111, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=939, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=\"5 + 3 = **8** and 7 + 2 = **9**. Now I'll multiply the two results.\\n\\n**Step 2: Multiply the results.**\", role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, caller={'type': 'direct'}, function=Function(arguments='{\"a\": 8, \"b\": 9}', name='multiply'), id='toolu_011dP6xwp1VHWRMYRLnHQH2v', type='function')], function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=111, prompt_tokens=939, total_tokens=1050, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=111, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
]
},
"metadata": {},
@@ -7361,12 +8325,12 @@
"- id: `chatcmpl-xxx`\n",
"- model: `claude-sonnet-4-6`\n",
"- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=43, prompt_tokens=1063, total_tokens=1106, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=43, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=1063, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
+ "- usage: `Usage(completion_tokens=43, prompt_tokens=1063, total_tokens=1106, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=43, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
"\n",
""
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='The final answer is:\\n\\n**(5 + 3) \\\\* (7 + 2) = 8 \\\\* 9 = 72** 🎉', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=43, prompt_tokens=1063, total_tokens=1106, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=43, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=1063, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='The final answer is:\\n\\n**(5 + 3) \\\\* (7 + 2) = 8 \\\\* 9 = 72** 🎉', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=43, prompt_tokens=1063, total_tokens=1106, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=43, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
]
},
"metadata": {},
@@ -7560,12 +8524,12 @@
"- id: `chatcmpl-xxx`\n",
"- model: `claude-sonnet-4-6`\n",
"- finish_reason: `tool_calls`\n",
- "- usage: `Usage(completion_tokens=267, prompt_tokens=809, total_tokens=1076, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=267, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=809, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
+ "- usage: `Usage(completion_tokens=267, prompt_tokens=809, total_tokens=1076, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=267, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
"\n",
""
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=\"## Available Tools\\n\\nI have the following tools available:\\n\\n1. **`simple_add`** – Adds two numbers together. Returns an integer.\\n2. **`multiply`** – Multiplies two numbers together. Returns an integer.\\n3. **`divide`** – Divides two numbers. Returns a number.\\n\\n---\\n\\n## Calculating ((10+5)*3)/(2+1)\\n\\nI'll break this down into steps:\\n- **Step 1:** `10 + 5` and `2 + 1` (these are independent, so I'll run them in parallel!)\\n\\n**Step 1: Parallel additions**\", role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, caller={'type': 'direct'}, function=Function(arguments='{\"a\": 10, \"b\": 5}', name='simple_add'), id='toolu_01G2P4J6NtF364CwAJpb75K1', type='function'), ChatCompletionMessageToolCall(index=2, caller={'type': 'direct'}, function=Function(arguments='{\"a\": 2, \"b\": 1}', name='simple_add'), id='toolu_01JDE9vTS1kNK2QbA47aUfac', type='function')], function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=267, prompt_tokens=809, total_tokens=1076, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=267, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=809, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=\"## Available Tools\\n\\nI have the following tools available:\\n\\n1. **`simple_add`** – Adds two numbers together. Returns an integer.\\n2. **`multiply`** – Multiplies two numbers together. Returns an integer.\\n3. **`divide`** – Divides two numbers. Returns a number.\\n\\n---\\n\\n## Calculating ((10+5)*3)/(2+1)\\n\\nI'll break this down into steps:\\n- **Step 1:** `10 + 5` and `2 + 1` (these are independent, so I'll run them in parallel!)\\n\\n**Step 1: Parallel additions**\", role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, caller={'type': 'direct'}, function=Function(arguments='{\"a\": 10, \"b\": 5}', name='simple_add'), id='toolu_01G2P4J6NtF364CwAJpb75K1', type='function'), ChatCompletionMessageToolCall(index=2, caller={'type': 'direct'}, function=Function(arguments='{\"a\": 2, \"b\": 1}', name='simple_add'), id='toolu_01JDE9vTS1kNK2QbA47aUfac', type='function')], function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=267, prompt_tokens=809, total_tokens=1076, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=267, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
]
},
"metadata": {},
@@ -7610,12 +8574,12 @@
"- id: `chatcmpl-xxx`\n",
"- model: `claude-sonnet-4-6`\n",
"- finish_reason: `tool_calls`\n",
- "- usage: `Usage(completion_tokens=112, prompt_tokens=1141, total_tokens=1253, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=112, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=1141, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
+ "- usage: `Usage(completion_tokens=112, prompt_tokens=1141, total_tokens=1253, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=112, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
"\n",
""
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=\"`10 + 5 = 15` and `2 + 1 = 3`. Now I'll multiply 15 × 3.\\n\\n**Step 2: Multiply**\", role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, caller={'type': 'direct'}, function=Function(arguments='{\"a\": 15, \"b\": 3}', name='multiply'), id='toolu_01XcWch985Y55eUaZXohNBwb', type='function')], function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=112, prompt_tokens=1141, total_tokens=1253, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=112, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=1141, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=\"`10 + 5 = 15` and `2 + 1 = 3`. Now I'll multiply 15 × 3.\\n\\n**Step 2: Multiply**\", role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, caller={'type': 'direct'}, function=Function(arguments='{\"a\": 15, \"b\": 3}', name='multiply'), id='toolu_01XcWch985Y55eUaZXohNBwb', type='function')], function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=112, prompt_tokens=1141, total_tokens=1253, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=112, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
]
},
"metadata": {},
@@ -7654,12 +8618,12 @@
"- id: `chatcmpl-xxx`\n",
"- model: `claude-sonnet-4-6`\n",
"- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=224, prompt_tokens=1285, total_tokens=1509, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=224, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=1285, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
+ "- usage: `Usage(completion_tokens=224, prompt_tokens=1285, total_tokens=1509, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=224, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
"\n",
""
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content=\"## Summary\\n\\nWe had a brief but productive session! Here's what we covered:\\n\\n1. **Listed the available tools** – We identified three math tools: `simple_add`, `multiply`, and `divide`.\\n\\n2. **Worked through a math expression** – We calculated **((10+5)*3)/(2+1)** step by step using the tools:\\n - `10 + 5 = 15` and `2 + 1 = 3` *(run in parallel)*\\n - `15 × 3 = 45`\\n - ⚠️ **We didn't quite finish!** The final step — dividing `45 ÷ 3` — was never completed. The final answer would be **15**, but the `divide` tool was never actually called to confirm it.\\n\\nSo we got about **90% of the way through** the calculation before wrapping up. Feel free to pick up where we left off next time! 😊\", role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=224, prompt_tokens=1285, total_tokens=1509, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=224, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=1285, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content=\"## Summary\\n\\nWe had a brief but productive session! Here's what we covered:\\n\\n1. **Listed the available tools** – We identified three math tools: `simple_add`, `multiply`, and `divide`.\\n\\n2. **Worked through a math expression** – We calculated **((10+5)*3)/(2+1)** step by step using the tools:\\n - `10 + 5 = 15` and `2 + 1 = 3` *(run in parallel)*\\n - `15 × 3 = 45`\\n - ⚠️ **We didn't quite finish!** The final step — dividing `45 ÷ 3` — was never completed. The final answer would be **15**, but the `divide` tool was never actually called to confirm it.\\n\\nSo we got about **90% of the way through** the calculation before wrapping up. Feel free to pick up where we left off next time! 😊\", role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=224, prompt_tokens=1285, total_tokens=1509, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=224, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
]
},
"metadata": {},
@@ -7687,8 +8651,7 @@
{
"data": {
"text/plain": [
- "[{'role': 'user',\n",
- " 'content': 'Tell me what tools you have available. Then calculate ((10+5)*3)/(2+1). ALWAYS use tools for math ops where available, and do tool calls in parallel where possible'},\n",
+ "['Tell me what tools you have available. Then calculate ((10+5)*3)/(2+1). ALWAYS use tools for math ops where available, and do tool calls in parallel where possible',\n",
" Message(content=\"## Available Tools\\n\\nI have the following tools available:\\n\\n1. **`simple_add`** – Adds two numbers together. Returns an integer.\\n2. **`multiply`** – Multiplies two numbers together. Returns an integer.\\n3. **`divide`** – Divides two numbers. Returns a number.\\n\\n---\\n\\n## Calculating ((10+5)*3)/(2+1)\\n\\nI'll break this down into steps:\\n- **Step 1:** `10 + 5` and `2 + 1` (these are independent, so I'll run them in parallel!)\\n\\n**Step 1: Parallel additions**\", role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, caller={'type': 'direct'}, function=Function(arguments='{\"a\": 10, \"b\": 5}', name='simple_add'), id='toolu_01G2P4J6NtF364CwAJpb75K1', type='function'), ChatCompletionMessageToolCall(index=2, caller={'type': 'direct'}, function=Function(arguments='{\"a\": 2, \"b\": 1}', name='simple_add'), id='toolu_01JDE9vTS1kNK2QbA47aUfac', type='function')], function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}),\n",
" {'tool_call_id': 'toolu_01G2P4J6NtF364CwAJpb75K1',\n",
" 'role': 'tool',\n",
@@ -7761,12 +8724,12 @@
"- id: `chatcmpl-xxx`\n",
"- model: `claude-sonnet-4-6`\n",
"- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=87, prompt_tokens=918, total_tokens=1005, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=87, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=918, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
+ "- usage: `Usage(completion_tokens=87, prompt_tokens=918, total_tokens=1005, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=87, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
"\n",
""
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content=\"Here's a summary of what was calculated so far:\\n\\n1. **1 + 2 = 3**\\n2. **3 + 2 = 5**\\n\\nHowever, the goal is **not yet complete**! We still need to perform one more calculation: **5 + 3**. Please send a follow-up message and I'll finish the calculation right away!\", role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=87, prompt_tokens=918, total_tokens=1005, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=87, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=918, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content=\"Here's a summary of what was calculated so far:\\n\\n1. **1 + 2 = 3**\\n2. **3 + 2 = 5**\\n\\nHowever, the goal is **not yet complete**! We still need to perform one more calculation: **5 + 3**. Please send a follow-up message and I'll finish the calculation right away!\", role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=87, prompt_tokens=918, total_tokens=1005, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=87, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
]
},
"execution_count": null,
@@ -7823,12 +8786,12 @@
"- id: `chatcmpl-xxx`\n",
"- model: `claude-sonnet-4-5-20250929`\n",
"- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=53, prompt_tokens=831, total_tokens=884, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=53, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=831, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None)`\n",
+ "- usage: `Usage(completion_tokens=53, prompt_tokens=831, total_tokens=884, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=53, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None)`\n",
"\n",
""
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-5-20250929', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='The result of adding 1 + 2 is **3**.\\n\\nThe tool_call_id I used was: **toolu_011YFz3D9hELtfe2faXFCru1**', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=53, prompt_tokens=831, total_tokens=884, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=53, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=831, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None))"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-5-20250929', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='The result of adding 1 + 2 is **3**.\\n\\nThe tool_call_id I used was: **toolu_011YFz3D9hELtfe2faXFCru1**', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=53, prompt_tokens=831, total_tokens=884, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=53, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None))"
]
},
"execution_count": null,
@@ -7909,12 +8872,12 @@
"- id: `chatcmpl-xxx`\n",
"- model: `claude-sonnet-4-5-20250929`\n",
"- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=64, prompt_tokens=1024, total_tokens=1088, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=64, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=1024, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None)`\n",
+ "- usage: `Usage(completion_tokens=64, prompt_tokens=1024, total_tokens=1088, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=64, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None)`\n",
"\n",
""
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-5-20250929', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='Perfect! I successfully:\\n1. Called `get_person` which returned Alice\\'s information (name: Alice, age: 30)\\n2. Passed that result to `greet_person` which greeted her with: \"Hello Alice, you are 30 years old!\"', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=64, prompt_tokens=1024, total_tokens=1088, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=64, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=1024, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None))"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-5-20250929', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='Perfect! I successfully:\\n1. Called `get_person` which returned Alice\\'s information (name: Alice, age: 30)\\n2. Passed that result to `greet_person` which greeted her with: \"Hello Alice, you are 30 years old!\"', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=64, prompt_tokens=1024, total_tokens=1088, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=64, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None))"
]
},
"execution_count": null,
@@ -8051,12 +9014,12 @@
"- id: `chatcmpl-xxx`\n",
"- model: `claude-sonnet-4-5-20250929`\n",
"- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=150, prompt_tokens=1119, total_tokens=1269, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=150, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=1119, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None)`\n",
+ "- usage: `Usage(completion_tokens=150, prompt_tokens=1119, total_tokens=1269, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=150, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None)`\n",
"\n",
""
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-5-20250929', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content=\"**Image Description:**\\nThis is an adorable photograph of a Cavalier King Charles Spaniel puppy. The puppy has the breed's characteristic coloring with a white face and chest, and rich brown/chestnut colored ears. The puppy is lying on grass and looking directly at the camera with large, expressive dark eyes and a sweet expression. In the background, there are purple/lavender colored flowers (possibly asters or similar blooms) that create a beautiful natural setting. The image has a warm, soft quality that perfectly captures the puppy's gentle and endearing nature.\\n\\n**Image Dimensions:**\\n- Width: 300 pixels\\n- Height: 200 pixels\", role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=150, prompt_tokens=1119, total_tokens=1269, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=150, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=1119, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None))"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-5-20250929', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content=\"**Image Description:**\\nThis is an adorable photograph of a Cavalier King Charles Spaniel puppy. The puppy has the breed's characteristic coloring with a white face and chest, and rich brown/chestnut colored ears. The puppy is lying on grass and looking directly at the camera with large, expressive dark eyes and a sweet expression. In the background, there are purple/lavender colored flowers (possibly asters or similar blooms) that create a beautiful natural setting. The image has a warm, soft quality that perfectly captures the puppy's gentle and endearing nature.\\n\\n**Image Dimensions:**\\n- Width: 300 pixels\\n- Height: 200 pixels\", role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=150, prompt_tokens=1119, total_tokens=1269, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=150, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None))"
]
},
"execution_count": null,
@@ -8143,12 +9106,12 @@
"- id: `chatcmpl-xxx`\n",
"- model: `claude-sonnet-4-5-20250929`\n",
"- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=68, prompt_tokens=948, total_tokens=1016, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=68, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=948, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None)`\n",
+ "- usage: `Usage(completion_tokens=68, prompt_tokens=948, total_tokens=1016, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=68, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None)`\n",
"\n",
""
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-5-20250929', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content=\"Perfect! I've successfully:\\n1. Called `get_config` which returned a configuration with `host: 'localhost'` and `port: 8080`\\n2. Passed that configuration to `use_config`, which processed it and returned `Host: localhost, Port: 8080`\", role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=68, prompt_tokens=948, total_tokens=1016, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=68, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=948, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None))"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-5-20250929', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content=\"Perfect! I've successfully:\\n1. Called `get_config` which returned a configuration with `host: 'localhost'` and `port: 8080`\\n2. Passed that configuration to `use_config`, which processed it and returned `Host: localhost, Port: 8080`\", role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=68, prompt_tokens=948, total_tokens=1016, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=68, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None))"
]
},
"execution_count": null,
@@ -8269,75 +9232,604 @@
},
{
"cell_type": "markdown",
- "id": "2ca77d9d",
- "metadata": {},
- "source": [
- "## Async"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "494116db",
- "metadata": {},
- "source": [
- "### AsyncChat"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "25921ffa",
+ "id": "6083fbf9",
"metadata": {},
"source": [
- "If you want to use LiteLLM in a webapp you probably want to use their async function `acompletion`.\n",
- "To make that easier we will implement our version of `AsyncChat` to complement it. It follows the same implementation as Chat as much as possible:"
+ "### Code Fence Tool"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "bb3811e0",
+ "id": "85f95676",
"metadata": {},
"outputs": [],
"source": [
- "#| export\n",
- "async def _alite_call_func(tc, tool_schemas, ns, tc_res=None, tc_res_eval=False):\n",
- " \"Call tool function asynchronously and return formatted result\"\n",
- " res = _call_func(tc, tool_schemas, ns, call_func_async, tc_res=tc_res)\n",
- " res = await maybe_await(res)\n",
- " return _mk_tool_result(tc, res, tc_res, tc_res_eval)"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "d0ec5da2",
- "metadata": {},
- "source": [
- "Testing the scenarios where the tool call was not in schemas:"
+ "def python(code: str):\n",
+ " \"Execute Python code and return stdout output\"\n",
+ " return 42"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "44cde8b8",
- "metadata": {},
- "outputs": [],
- "source": [
- "result = await _alite_call_func(fake_tc, [toolsc], globals())\n",
- "test_eq(result['content'], \"Tool not defined in tool_schemas: hallucinated_tool\")"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "2cf71e26",
+ "id": "770688f4",
"metadata": {},
- "source": [
- "or schemas was missing...:"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "dda2bc30",
+ "outputs": [
+ {
+ "data": {
+ "text/markdown": [
+ "\n",
+ "\n",
+ "**gemini/gemini-3-pro-preview:**\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(**gemini/gemini-3-pro-preview:**)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "\n",
+ "\n",
+ "I will write a Python function to generate a random number between 1 and 100, call it, and print the result. \n",
+ "\n",
+ "Here is the code that will be executed:\n",
+ "\n",
+ "`````py\n",
+ "import random\n",
+ "\n",
+ "def generate_random_number(min_val=1, max_val=100):\n",
+ " \"\"\"Generates a random integer between min_val and max_val (inclusive).\"\"\"\n",
+ " return random.randint(min_val, max_val)\n",
+ "\n",
+ "# Call statement\n",
+ "random_num = generate_random_number()\n",
+ "print(random_num)\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "The execution is complete! The randomly generated number is **42**.\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(I will write a Python function to generate a random number between 1 and 100, call it, and print the result. \n",
+ "\n",
+ "Here is the code that will be executed:\n",
+ "\n",
+ "`````py\n",
+ "import random\n",
+ "\n",
+ "def generate_random_number(min_val=1, max_val=100):\n",
+ " \"\"\"Generates a random integer between min_val and max_val (inclusive).\"\"\"\n",
+ " return random.randint(min_val, max_val)\n",
+ "\n",
+ "# Call statement\n",
+ "random_num = generate_random_number()\n",
+ "print(random_num)\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "The execution is complete! The randomly generated number is **42**.)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "\n",
+ "\n",
+ "`````py\n",
+ "import random\n",
+ "print(random.randint(1, 100))\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "The execution is complete! The randomly generated number this time is also **42**! (What a coincidence!).\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(`````py\n",
+ "import random\n",
+ "print(random.randint(1, 100))\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "The execution is complete! The randomly generated number this time is also **42**! (What a coincidence!).)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "\n",
+ "\n",
+ "**gemini/gemini-3-flash-preview:**\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(**gemini/gemini-3-flash-preview:**)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "\n",
+ "\n",
+ "Here is the code to generate a random number between 1 and 100.\n",
+ "\n",
+ "`````py\n",
+ "import random\n",
+ "\n",
+ "def generate_random_number():\n",
+ " return random.randint(1, 100)\n",
+ "\n",
+ "print(generate_random_number())\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "The random number generated is 42.\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(Here is the code to generate a random number between 1 and 100.\n",
+ "\n",
+ "`````py\n",
+ "import random\n",
+ "\n",
+ "def generate_random_number():\n",
+ " return random.randint(1, 100)\n",
+ "\n",
+ "print(generate_random_number())\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "The random number generated is 42.)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "\n",
+ "\n",
+ "`````py\n",
+ "import random\n",
+ "\n",
+ "def generate_random_number():\n",
+ " return random.randint(1, 100)\n",
+ "\n",
+ "print(generate_random_number())\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "`````py\n",
+ "import random\n",
+ "\n",
+ "def generate_random_number():\n",
+ " return random.randint(1, 100)\n",
+ "\n",
+ "print(generate_random_number())\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "The random number generated is 42.\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(`````py\n",
+ "import random\n",
+ "\n",
+ "def generate_random_number():\n",
+ " return random.randint(1, 100)\n",
+ "\n",
+ "print(generate_random_number())\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "`````py\n",
+ "import random\n",
+ "\n",
+ "def generate_random_number():\n",
+ " return random.randint(1, 100)\n",
+ "\n",
+ "print(generate_random_number())\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "The random number generated is 42.)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "\n",
+ "\n",
+ "**claude-sonnet-4-6:**\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(**claude-sonnet-4-6:**)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "\n",
+ "\n",
+ "\n",
+ "`````py\n",
+ "import random\n",
+ "\n",
+ "def random_number_generator(start=1, end=100):\n",
+ " return random.randint(start, end)\n",
+ "\n",
+ "print(random_number_generator())\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠🧠🧠🧠🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "The randomly generated number is **42**! 🎲\n",
+ "\n",
+ "Here's a quick breakdown of the function:\n",
+ "- **`random_number_generator(start=1, end=100)`** — accepts a range with default values of `1` to `100`.\n",
+ "- It uses Python's built-in **`random.randint(start, end)`** to generate a random integer **inclusive** of both endpoints.\n",
+ "- You can customize the range, e.g., `random_number_generator(1, 500)` to generate a number between 1 and 500.\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(\n",
+ "`````py\n",
+ "import random\n",
+ "\n",
+ "def random_number_generator(start=1, end=100):\n",
+ " return random.randint(start, end)\n",
+ "\n",
+ "print(random_number_generator())\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠🧠🧠🧠🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "The randomly generated number is **42**! 🎲\n",
+ "\n",
+ "Here's a quick breakdown of the function:\n",
+ "- **`random_number_generator(start=1, end=100)`** — accepts a range with default values of `1` to `100`.\n",
+ "- It uses Python's built-in **`random.randint(start, end)`** to generate a random integer **inclusive** of both endpoints.\n",
+ "- You can customize the range, e.g., `random_number_generator(1, 500)` to generate a number between 1 and 500.)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "\n",
+ "\n",
+ "`````py\n",
+ "import random\n",
+ "\n",
+ "def random_number_generator(start=1, end=100):\n",
+ " return random.randint(start, end)\n",
+ "\n",
+ "print(random_number_generator())\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "The randomly generated number is **42** again! 🎲🎉\n",
+ "\n",
+ "What are the odds! We got **42** twice in a row! That's the beauty (and sometimes funny coincidence) of random number generation. Run it again if you'd like to see a different number! 😄\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(`````py\n",
+ "import random\n",
+ "\n",
+ "def random_number_generator(start=1, end=100):\n",
+ " return random.randint(start, end)\n",
+ "\n",
+ "print(random_number_generator())\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "The randomly generated number is **42** again! 🎲🎉\n",
+ "\n",
+ "What are the odds! We got **42** twice in a row! That's the beauty (and sometimes funny coincidence) of random number generation. Run it again if you'd like to see a different number! 😄)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "\n",
+ "\n",
+ "**openai/gpt-5.4:**\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(**openai/gpt-5.4:**)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "\n",
+ "\n",
+ "`````py\n",
+ "import random\n",
+ "\n",
+ "def rng():\n",
+ " return random.random()\n",
+ "\n",
+ "print(rng())\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "42\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(`````py\n",
+ "import random\n",
+ "\n",
+ "def rng():\n",
+ " return random.random()\n",
+ "\n",
+ "print(rng())\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "42)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "\n",
+ "\n",
+ "`````py\n",
+ "import random\n",
+ "\n",
+ "def rng():\n",
+ " return random.random()\n",
+ "\n",
+ "print(rng())\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "42\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(`````py\n",
+ "import random\n",
+ "\n",
+ "def rng():\n",
+ " return random.random()\n",
+ "\n",
+ "print(rng())\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "42)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "cft_pr = 'Write a random number generator function and print a call statement which will be executed as a code fence tool. Tell me the number once you get the execution result'\n",
+ "for m in ms:\n",
+ " display(Markdown(f'**{m}:**'))\n",
+ " chat = Chat(m, sp=cft_sp, tools=[python])\n",
+ " res = chat(cft_pr, stream=True)\n",
+ " fmt = display_stream(res)\n",
+ " try: test(fmt.outp, '`````py', operator.contains)\n",
+ " except AssertionError: display(Markdown(f'**Model {m} not following instructions**')); continue\n",
+ " test(fmt.outp, '`````result\\n42\\n`````', operator.contains)\n",
+ " res = chat(\"Do one more time\", stream=True)\n",
+ " fmt = display_stream(res)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "2ca77d9d",
+ "metadata": {},
+ "source": [
+ "## Async"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "494116db",
+ "metadata": {},
+ "source": [
+ "### AsyncChat"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "25921ffa",
+ "metadata": {},
+ "source": [
+ "If you want to use LiteLLM in a webapp you probably want to use their async function `acompletion`.\n",
+ "To make that easier we will implement our version of `AsyncChat` to complement it. It follows the same implementation as Chat as much as possible:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "bb3811e0",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "#| export\n",
+ "async def _alite_call_func(tc, tool_schemas, ns, tc_res=None, tc_res_eval=False):\n",
+ " \"Call tool function asynchronously and return formatted result\"\n",
+ " res = _call_func(tc, tool_schemas, ns, call_func_async, tc_res=tc_res)\n",
+ " res = await maybe_await(res)\n",
+ " return _mk_tool_result(tc, res, tc_res, tc_res_eval)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "d0ec5da2",
+ "metadata": {},
+ "source": [
+ "Testing the scenarios where the tool call was not in schemas:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "44cde8b8",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "result = await _alite_call_func(fake_tc, [toolsc], globals())\n",
+ "test_eq(result['content'], \"Tool not defined in tool_schemas: hallucinated_tool\")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "2cf71e26",
+ "metadata": {},
+ "source": [
+ "or schemas was missing...:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "dda2bc30",
"metadata": {},
"outputs": [],
"source": [
@@ -8354,13 +9846,40 @@
"source": [
"#| export\n",
"@asave_iter\n",
- "async def astream_with_complete(self, agen, postproc=noop):\n",
+ "async def astream_with_complete(self, agen, postproc=noop, stop_callables=None):\n",
" chunks = []\n",
+ " stop = False\n",
" async for chunk in agen:\n",
" chunks.append(chunk)\n",
" postproc(chunk)\n",
- " yield chunk\n",
- " self.value = stream_chunk_builder(chunks)"
+ " if stop_callables and not stop:\n",
+ " for f in stop_callables:\n",
+ " txt = completion_text(chunks)\n",
+ " if res:=f(txt):\n",
+ " chunk, final_txt = _trim_chunk(chunk, txt, res)\n",
+ " stop = True\n",
+ " yield chunk\n",
+ " if stop: yield mk_stream_chunk(reasoning_content=\"processing\")\n",
+ " else: yield chunk\n",
+ " res = stream_chunk_builder(chunks)\n",
+ " if stop: res.choices[0].message.content = final_txt\n",
+ " self.value = res"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "cd9de8ec",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "#| export\n",
+ "async def run_fence_tool_async(lang, code, ns):\n",
+ " \"Run the mapped tool for `lang` with the code, return result fence\"\n",
+ " tname = _lang2tool[lang]\n",
+ " arg = dict(code=code) if lang == 'py' else dict(command=code)\n",
+ " res = await call_func_async(tname, arg, ns=ns, raise_on_err=False)\n",
+ " return _mk_result_fence(_trunc_str(str(res)))"
]
},
{
@@ -8380,18 +9899,18 @@
"source": [
"#| export\n",
"class AsyncChat(Chat):\n",
- " async def _call(self, msg=None, prefill=None, temp=None, think=None, search=None, stream=False, max_steps=2, step=1,\n",
- " final_prompt=None, tool_choice=None, max_tokens=None, n_workers=8, pause=0.001, tc_timeout=7200, **kwargs):\n",
+ " async def _call(self, msg=None, prefill=None, temp=None, think=None, search=None, stream=False, max_steps=2, step=1, final_prompt=None, tool_choice=None, max_tokens=None, stop_callables=None, n_workers=8, pause=0.001, tc_timeout=7200, **kwargs):\n",
" if step>max_steps+1: return\n",
" if self.callkw: kwargs = {**self.callkw, **kwargs}\n",
" prefill, max_tokens = self._prep_call(prefill, search, max_tokens, kwargs, stream=stream, think=think)\n",
+ " stop_callables = listify(stop_callables) + listify(kwargs.pop('stop_callables', None))\n",
" mt = {} if max_tokens in (None,0) else dict(max_tokens=int(max_tokens))\n",
- " res = await self.completefunc(model=self.model, messages=self._prep_msg(msg, prefill), stream=stream, num_retries=2,\n",
+ " res = await self.completefunc(model=self.model, messages=self._prep_msg(msg, prefill, stop_callables), stream=stream, num_retries=2,\n",
" tools=self.tool_schemas, tool_choice=tool_choice, **mt,\n",
" temperature=None if think else ifnone(temp,self.temp), caching=self.cache and 'claude' not in self.model, **kwargs)\n",
" if stream:\n",
" if prefill: yield _mk_prefill(prefill)\n",
- " res = astream_with_complete(res,postproc=cite_footnote)\n",
+ " res = astream_with_complete(res,postproc=cite_footnote,stop_callables=stop_callables)\n",
" async for chunk in res: yield chunk\n",
" res = res.value\n",
" elif not isinstance(res, ModelResponse):\n",
@@ -8427,7 +9946,19 @@
" if len(t['content'])>1000: t['content'] = _cwe_msg + _trunc_str(t['content'], mx=1000)\n",
" async for result in self._call(\n",
" prompt, prefill, temp, think, search, stream, max_steps, step+1,\n",
- " final_prompt, tool_choice='none', **kwargs): yield result"
+ " final_prompt, tool_choice='none', **kwargs): yield result\n",
+ " elif (langs := _active_fence_langs(self.tool_schemas)):\n",
+ " m = self.hist[-1]\n",
+ " if m.role == 'assistant':\n",
+ " if fence := extract_fence_call(m.content or ''):\n",
+ " lang, code = fence\n",
+ " out = await run_fence_tool_async(lang, code, self.ns)\n",
+ " m.content += out \n",
+ " if stream: yield mk_stream_chunk(content=out, role='assistant')\n",
+ " if step <= max_steps:\n",
+ " async for result in self._call(\n",
+ " None, prefill, temp, think, search, stream, max_steps, step+1,\n",
+ " final_prompt, tool_choice, **kwargs): yield result"
]
},
{
@@ -8613,12 +10144,12 @@
"- id: `chatcmpl-xxx`\n",
"- model: `claude-opus-4-5-20251101`\n",
"- finish_reason: `content_filter`\n",
- "- usage: `Usage(completion_tokens=4, prompt_tokens=30, total_tokens=34, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=4, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=30, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None)`\n",
+ "- usage: `Usage(completion_tokens=4, prompt_tokens=30, total_tokens=34, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=4, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None)`\n",
"\n",
""
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-opus-4-5-20251101', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='content_filter', index=0, message=Message(content='AI server provider content filter was applied to this request. ', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=4, prompt_tokens=30, total_tokens=34, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=4, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=30, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None))"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-opus-4-5-20251101', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='content_filter', index=0, message=Message(content='AI server provider content filter was applied to this request. ', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=4, prompt_tokens=30, total_tokens=34, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=4, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None))"
]
},
"execution_count": null,
@@ -8660,212 +10191,310 @@
},
{
"cell_type": "markdown",
- "id": "2150a316",
+ "id": "c2c12c11",
"metadata": {},
"source": [
- "This is what our outputs look like with streaming results:"
+ "Tools can return `StopResponse` to enforce the tool loop stops immediately."
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "57698c63",
+ "id": "dacc8a95",
"metadata": {},
"outputs": [
{
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Sure! Let me calculate that for you right away.\n",
- "🔧 async_add\n"
- ]
+ "data": {
+ "text/markdown": [
+ "Sure! I'll start by calling `stop_tool` with `'halt'` first, and then call `simple_add(1, 2)` afterward.\n",
+ "\n",
+ "**Step 1: Calling `stop_tool` with `'halt'`**\n",
+ "\n",
+ "🔧 stop_tool({\"msg\": \"halt\"})\n",
+ "\n",
+ "\n",
+ "\n",
+ "\n",
+ "- id: `chatcmpl-xxx`\n",
+ "- model: `claude-sonnet-4-6`\n",
+ "- finish_reason: `tool_calls`\n",
+ "- usage: `Usage(completion_tokens=110, prompt_tokens=707, total_tokens=817, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=110, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
+ "\n",
+ " "
+ ],
+ "text/plain": [
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=\"Sure! I'll start by calling `stop_tool` with `'halt'` first, and then call `simple_add(1, 2)` afterward.\\n\\n**Step 1: Calling `stop_tool` with `'halt'`**\", role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, caller={'type': 'direct'}, function=Function(arguments='{\"msg\": \"halt\"}', name='stop_tool'), id='toolu_01EmeuvV8WeA9KqArDvEAz48', type='function')], function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=110, prompt_tokens=707, total_tokens=817, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=110, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
},
{
"data": {
"text/plain": [
- "{'tool_call_id': 'toolu_01KTpKieSog8ChHBbYNFd6Ce',\n",
+ "{'tool_call_id': 'toolu_01EmeuvV8WeA9KqArDvEAz48',\n",
" 'role': 'tool',\n",
- " 'name': 'async_add',\n",
- " 'content': '12'}"
+ " 'name': 'stop_tool',\n",
+ " 'content': 'Can not continue: halt'}"
]
},
"metadata": {},
"output_type": "display_data"
},
{
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "The result of **5 + 7 = 12**! 🎉"
- ]
+ "data": {
+ "text/markdown": [
+ "Here's a summary of what happened:\n",
+ "\n",
+ "- ✅ **`stop_tool('halt')`** was called successfully and returned: `\"Can not continue: halt\"`\n",
+ "- ⏳ **`simple_add(1, 2)`** was **not yet called** — I ran out of tool calls for this turn.\n",
+ "\n",
+ "**Further work needed:** On the next message, I will call `simple_add(1, 2)` to complete the second step. Please prompt me to continue!\n",
+ "\n",
+ "\n",
+ "\n",
+ "- id: `chatcmpl-xxx`\n",
+ "- model: `claude-sonnet-4-6`\n",
+ "- finish_reason: `stop`\n",
+ "- usage: `Usage(completion_tokens=114, prompt_tokens=884, total_tokens=998, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=114, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
+ "\n",
+ " "
+ ],
+ "text/plain": [
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='Here\\'s a summary of what happened:\\n\\n- ✅ **`stop_tool(\\'halt\\')`** was called successfully and returned: `\"Can not continue: halt\"`\\n- ⏳ **`simple_add(1, 2)`** was **not yet called** — I ran out of tool calls for this turn.\\n\\n**Further work needed:** On the next message, I will call `simple_add(1, 2)` to complete the second step. Please prompt me to continue!', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=114, prompt_tokens=884, total_tokens=998, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=114, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
}
],
"source": [
- "chat_with_tools = AsyncChat(model, tools=[async_add])\n",
- "res = await chat_with_tools(\"What is 5 + 7? Use the tool to calculate it.\", stream=True)\n",
- "async for o in res:\n",
- " if isinstance(o,ModelResponseStream): print(delta_text(o) or '',end='')\n",
- " elif isinstance(o,dict): _display(o)"
+ "def stop_tool(msg: str) -> str:\n",
+ " \"A tool that stops the loop\"\n",
+ " return StopResponse(f\"Can not continue: {msg}\")\n",
+ "\n",
+ "chat = Chat(model, tools=[simple_add, stop_tool])\n",
+ "res = chat(\"First call stop_tool with 'halt', then call simple_add(1,2). Use both tools, one after the other (not at the same time).\", max_steps=10, return_all=True)\n",
+ "# Should only have 1 round of tool calls + final response, never reaching simple_add in a second round\n",
+ "for r in res: _display(r)"
]
},
{
- "cell_type": "markdown",
- "id": "f309f775",
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "7a6199ff",
"metadata": {},
+ "outputs": [],
"source": [
- "Here's a complete `ModelResponse` taken from the response stream:"
+ "#| export\n",
+ "class AsyncStreamFormatter(StreamFormatter):\n",
+ " async def format_stream(self, rs):\n",
+ " \"Format the response stream for markdown display.\"\n",
+ " async for o in rs: yield self.format_item(o)"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "5203c123",
+ "id": "f627276d",
"metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-5', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=\"I'll calculate ((10 + 5) * 3) / (2 + 1) step by step:\", role='assistant', tool_calls=[ChatCompletionMessageToolCall(function=Function(arguments='{\"a\": 10, \"b\": 5}', name='simple_add'), id='toolu_018BGyenjiRkDQFU1jWP6qRo', type='function'), ChatCompletionMessageToolCall(function=Function(arguments='{\"a\": 2, \"b\": 1}', name='simple_add'), id='toolu_01CWqrNQvoRjf1Q1GLpTUgQR', type='function')], function_call=None, provider_specific_fields=None))], usage=Usage(completion_tokens=228, prompt_tokens=794, total_tokens=1022, completion_tokens_details=None, prompt_tokens_details=None))\n"
- ]
- }
- ],
+ "outputs": [],
"source": [
- "resp = ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-5', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=\"I'll calculate ((10 + 5) * 3) / (2 + 1) step by step:\", role='assistant', tool_calls=[ChatCompletionMessageToolCall(function=Function(arguments='{\"a\": 10, \"b\": 5}', name='simple_add'), id='toolu_018BGyenjiRkDQFU1jWP6qRo', type='function'), ChatCompletionMessageToolCall(function=Function(arguments='{\"a\": 2, \"b\": 1}', name='simple_add'), id='toolu_01CWqrNQvoRjf1Q1GLpTUgQR', type='function')], function_call=None, provider_specific_fields=None))], usage=Usage(completion_tokens=228, prompt_tokens=794, total_tokens=1022, prompt_tokens_details=None))\n",
- "print(repr(resp))"
+ "mock_tool_call = ChatCompletionMessageToolCall(\n",
+ " id=\"toolu_123abc456def\", type=\"function\", \n",
+ " function=Function( name=\"simple_add\", arguments='{\"a\": 5, \"b\": 3}' )\n",
+ ")\n",
+ "\n",
+ "mock_response = ModelResponse(usage=Usage(prompt_tokens=0, completion_tokens=0, total_tokens=0), model=haik45)\n",
+ "mock_response.choices = [type('Choice', (), {\n",
+ " 'message': type('Message', (), {\n",
+ " 'tool_calls': [mock_tool_call]\n",
+ " })()\n",
+ "})()]\n",
+ "\n",
+ "mock_tool_result = {\n",
+ " 'tool_call_id': mock_tool_call.id, 'role': 'tool', \n",
+ " 'name': 'simple_add', 'content': '8'\n",
+ "}"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "bc69f062",
+ "id": "f9362bb7",
"metadata": {},
"outputs": [
{
- "data": {
- "text/plain": [
- "ChatCompletionMessageToolCall(function=Function(arguments='{\"a\": 10, \"b\": 5}', name='simple_add'), id='toolu_018BGyenjiRkDQFU1jWP6qRo', type='function')"
- ]
- },
- "execution_count": null,
- "metadata": {},
- "output_type": "execute_result"
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "\n",
+ "- ⏳ simple_add(a=5, b=3) ⏳\n",
+ "---\n",
+ "\n",
+ "\n",
+ "\n",
+ "simple_add(a=5, b=3)→8 \n",
+ "\n",
+ "\n"
+ ]
}
],
"source": [
- "tc=resp.choices[0].message.tool_calls[0]\n",
- "tc"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "3a0c0add",
- "metadata": {},
- "outputs": [],
- "source": [
- "tr={'tool_call_id': 'toolu_018BGyenjiRkDQFU1jWP6qRo', 'role': 'tool','name': 'simple_add',\n",
- " 'content': '15 is the answer! ' +'.'*2000}"
+ "fmt = AsyncStreamFormatter()\n",
+ "print(fmt.format_item(mock_response))\n",
+ "print('---')\n",
+ "print(fmt.format_item(mock_tool_result))"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "049f141f",
+ "id": "d7f3452b",
"metadata": {},
"outputs": [],
"source": [
"#| export\n",
- "def trunc_param(v, mx=40):\n",
- " \"Truncate and escape param value for display\"\n",
- " tp = _trunc_str(str(v).replace('`', r'\\`'), mx=mx, replace=None, skip=0)\n",
- " try: return ast.literal_eval(tp)\n",
- " except Exception: return repr(tp).replace('\\\\\\\\', '\\\\')\n",
- "\n",
- "def _tc_summary(tc, tr=None):\n",
- " \"Format tool call as func(params) → result string\"\n",
- " args = json.loads(tc.function.arguments)\n",
- " params = ', '.join(f\"{k}={trunc_param(v)}\" for k,v in args.items())\n",
- " res = f\"→{trunc_param(tr.get('content',''))}\" if tr else ''\n",
- " return ''+escape(f\"{tc.function.name}({params}){res}\")+''\n",
- "\n",
- "def _trunc_content(content, mx):\n",
- " \"Truncate tool result content, respecting '_full' flag\"\n",
- " if isinstance(content, dict) and '_full' in content and len(content)==1: return content['_full']\n",
- " return _trunc_str(content, mx=mx)\n",
- "\n",
- "def mk_tr_details(tr, tc, mx=2000):\n",
- " \"Create block for tool call as JSON\"\n",
- " args = {k:_trunc_str(v, mx=mx*5) for k,v in json.loads(tc.function.arguments).items()}\n",
- " res = {'id':tr['tool_call_id'],\n",
- " 'call':{'function': tc.function.name, 'arguments': args},\n",
- " 'result':_trunc_content(tr.get('content'), mx=mx),}\n",
- " summ = f\"{_tc_summary(tc,tr)} \"\n",
- " return f\"\\n\\n{tool_dtls_tag}\\n{summ}\\n\\n```json\\n{dumps(res, indent=2)}\\n```\\n\\n \\n\\n\""
+ "@delegates(AsyncStreamFormatter)\n",
+ "async def adisplay_stream(rs, **kwargs):\n",
+ " \"Use IPython.display to markdown display the response stream.\"\n",
+ " try: from IPython.display import display, Markdown\n",
+ " except ModuleNotFoundError: raise ModuleNotFoundError(\"This function requires ipython. Please run `pip install ipython` to use.\")\n",
+ " fmt = AsyncStreamFormatter(**kwargs)\n",
+ " md,h = '',display(Markdown(' '), display_id=True)\n",
+ " async for o in fmt.format_stream(rs):\n",
+ " md += o\n",
+ " if md: h.update(Markdown(md))\n",
+ " return fmt"
]
},
{
- "cell_type": "code",
- "execution_count": null,
- "id": "8cb7f078",
+ "cell_type": "markdown",
+ "id": "bcf74cd3",
"metadata": {},
- "outputs": [],
"source": [
- "#| export\n",
- "status_re = re.compile(r'^- ⏳ (.*) ⏳$|^🧠+$', re.MULTILINE)\n",
- "\n",
- "class StreamFormatter:\n",
- " def __init__(self, mx=2000, debug=False, showthink=False):\n",
- " self.outp,self.tcs = '',{}\n",
- " store_attr()\n",
- " \n",
- " def format_item(self, o):\n",
- " \"Format a single item from the response stream.\"\n",
- " res = ''\n",
- " if self.debug: print(o)\n",
- " if isinstance(o, ModelResponseStream):\n",
- " d = o.choices[0].delta\n",
- " if nested_idx(d, 'reasoning_content') and d['reasoning_content']!='{\"text\": \"\"}':\n",
- " if self.showthink: res += str(nested_idx(d, 'reasoning_content'))\n",
- " res+= '🧠' if not self.outp or self.outp[-1]=='🧠' else '\\n\\n🧠'\n",
- " elif self.outp and self.outp[-1] == '🧠': res+= '\\n\\n'\n",
- " if c:=d.content: res+=f\"\\n\\n{c}\" if res and res[-1] == '🧠' else c\n",
- " for img in getattr(d, 'images', []): res += f\"\\n\\n})\\n\\n\"\n",
- " elif isinstance(o, ModelResponse):\n",
- " if c:=getattr(contents(o),'tool_calls',None):\n",
- " self.tcs = {tc.id:tc for tc in c}\n",
- " for tc in c: res += f\"\\n- ⏳ {_tc_summary(tc)} ⏳\"\n",
- " elif isinstance(o, dict) and 'tool_call_id' in o:\n",
- " res += mk_tr_details(o, self.tcs.pop(o['tool_call_id']), mx=self.mx)\n",
- " self.outp+=res\n",
- " return res\n",
- " \n",
- " def format_stream(self, rs):\n",
- " \"Format the response stream for markdown display.\"\n",
- " for o in rs: yield self.format_item(o)"
+ "Test of workaround/fix for https://github.com/BerriAI/litellm/issues/25869 :"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "a242a545",
+ "id": "c1849766",
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "data": {
+ "text/markdown": [
+ "\n",
+ "\n",
+ "\n",
+ "- ⏳ simple_add(b=1, a=1) ⏳\n",
+ "\n",
+ "\n",
+ "simple_add(b=1, a=1)→2 \n",
+ "\n",
+ "\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(\n",
+ "- ⏳ simple_add(b=1, a=1) ⏳\n",
+ "\n",
+ "\n",
+ "simple_add(b=1, a=1)→2 \n",
+ "\n",
+ ")"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
"source": [
- "stream_msg = ModelResponseStream([StreamingChoices(delta=Delta(content=\"Hello world!\"))])\n",
- "sf = StreamFormatter().format_item(stream_msg)"
+ "chat = AsyncChat(ms[1], tools=[simple_add], search='l')\n",
+ "# This prompt forces Turn 1: Search + simple_add, then Turn 2: Summary\n",
+ "pr = \"As a test of tool calling, at the same time search the web for Brisbane's population and also use simple_add to add 1+1; after you have the results for both, tell me a joke!\"\n",
+ "\n",
+ "res = await chat(pr, stream=True, max_steps=5)\n",
+ "fmt = await adisplay_stream(res)"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "e50c01e9",
+ "id": "e0d7e91d",
"metadata": {},
"outputs": [
{
"data": {
+ "text/markdown": [
+ "The population of Brisbane is currently estimated to be approximately **2,599,740** (around 2.6 million) for the metro area in 2026. Meanwhile, adding 1 + 1 using my internal tools gives a result of **2**.\n",
+ "\n",
+ "And as promised, here is a joke for you:\n",
+ "\n",
+ "**Why don't scientists trust atoms?**\n",
+ "Because they make up everything!\n",
+ "\n",
+ "\n",
+ "\n",
+ "- id: `chatcmpl-xxx`\n",
+ "- model: `gemini-3-flash-preview`\n",
+ "- finish_reason: `stop`\n",
+ "- usage: `Usage(completion_tokens=256, prompt_tokens=225, total_tokens=625, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=167, rejected_prediction_tokens=None, text_tokens=89, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=225, image_tokens=None, video_tokens=None), cache_read_input_tokens=None)`\n",
+ "\n",
+ " "
+ ],
"text/plain": [
- "'🧠'"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-flash-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content=\"The population of Brisbane is currently estimated to be approximately **2,599,740** (around 2.6 million) for the metro area in 2026. Meanwhile, adding 1 + 1 using my internal tools gives a result of **2**.\\n\\nAnd as promised, here is a joke for you:\\n\\n**Why don't scientists trust atoms?**\\nBecause they make up everything!\", role='assistant', tool_calls=None, function_call=None, images=[], thinking_blocks=[], provider_specific_fields={'thought_signatures': ['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', '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', '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'], 'server_side_tool_invocations': [{'tool_type': 'GOOGLE_SEARCH_WEB', 'id': 'n1h9jed0', 'args': {'queries': ['current population of Brisbane 2026']}, 'thought_signature': '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', 'response': {'search_suggestions': '\\n\\n
\\n
\\n \\n \\n \\n \\n \\n
\\n \\n \\n \\n \\n \\n \\n
\\n
\\n
\\n
\\n
\\n
\\n \\n \\n \\n \\n \\n
\\n \\n \\n \\n \\n \\n \\n
\\n
\\n
\\n
\n",
"\n",
- "**Step 1: Calling `stop_tool` with `'halt'`**\n",
+ "Sure! Let me calculate that for you right away.\n",
+ "- ⏳ simple_add(a=5, b=7) ⏳\n",
"\n",
- "🔧 stop_tool({\"msg\": \"halt\"})\n",
+ "\n",
+ "simple_add(a=5, b=7)→12\n",
+ " \n",
"\n",
- "- id: `chatcmpl-xxx`\n",
- "- model: `claude-sonnet-4-6`\n",
- "- finish_reason: `tool_calls`\n",
- "- usage: `Usage(completion_tokens=110, prompt_tokens=707, total_tokens=817, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=110, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=707, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
+ "The result of **5 + 7 = 12**. 🎉\n",
"\n",
- " "
+ "
"
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=\"Sure! I'll start by calling `stop_tool` with `'halt'` first, and then call `simple_add(1, 2)` afterward.\\n\\n**Step 1: Calling `stop_tool` with `'halt'`**\", role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=1, caller={'type': 'direct'}, function=Function(arguments='{\"msg\": \"halt\"}', name='stop_tool'), id='toolu_01EmeuvV8WeA9KqArDvEAz48', type='function')], function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=110, prompt_tokens=707, total_tokens=817, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=110, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=707, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "text/plain": [
- "{'tool_call_id': 'toolu_01EmeuvV8WeA9KqArDvEAz48',\n",
- " 'role': 'tool',\n",
- " 'name': 'stop_tool',\n",
- " 'content': 'Can not continue: halt'}"
+ "Markdown(Sure! Let me calculate that for you right away.\n",
+ "- ⏳ simple_add(a=5, b=7) ⏳\n",
+ "\n",
+ "\n",
+ "simple_add(a=5, b=7)→12 \n",
+ "\n",
+ "The result of **5 + 7 = 12**. 🎉)"
]
},
"metadata": {},
"output_type": "display_data"
- },
+ }
+ ],
+ "source": [
+ "chat = Chat(model, tools=[simple_add])\n",
+ "res = chat(\"What is 5 + 7? Use the tool to calculate it.\", stream=True)\n",
+ "fmt = display_stream(res)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "9ef4c620",
+ "metadata": {},
+ "outputs": [
{
"data": {
"text/markdown": [
- "Here's a summary of what happened:\n",
+ "\n",
"\n",
- "- ✅ **`stop_tool('halt')`** was called successfully and returned: `\"Can not continue: halt\"`\n",
- "- ⏳ **`simple_add(1, 2)`** was **not yet called** — I ran out of tool calls for this turn.\n",
+ "Sure! Let me calculate that for you right away.\n",
+ "- ⏳ async_add(a=5, b=7) ⏳\n",
"\n",
- "**Further work needed:** On the next message, I will call `simple_add(1, 2)` to complete the second step. Please prompt me to continue!\n",
+ "\n",
+ "async_add(a=5, b=7)→12\n",
+ "```json\n",
+ "{\n",
+ " \"id\": \"toolu_01KTpKieSog8ChHBbYNFd6Ce\",\n",
+ " \"call\": {\n",
+ " \"function\": \"async_add\",\n",
+ " \"arguments\": {\n",
+ " \"a\": \"5\",\n",
+ " \"b\": \"7\"\n",
+ " }\n",
+ " },\n",
+ " \"result\": \"12\"\n",
+ "}\n",
+ "```\n",
"\n",
- "- id: `chatcmpl-xxx`\n",
- "- model: `claude-sonnet-4-6`\n",
- "- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=114, prompt_tokens=884, total_tokens=998, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=114, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=884, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
+ " \n",
"\n",
- " "
+ "The result of **5 + 7 = 12**! 🎉\n",
+ "\n",
+ "
"
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='Here\\'s a summary of what happened:\\n\\n- ✅ **`stop_tool(\\'halt\\')`** was called successfully and returned: `\"Can not continue: halt\"`\\n- ⏳ **`simple_add(1, 2)`** was **not yet called** — I ran out of tool calls for this turn.\\n\\n**Further work needed:** On the next message, I will call `simple_add(1, 2)` to complete the second step. Please prompt me to continue!', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=114, prompt_tokens=884, total_tokens=998, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=114, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=884, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
+ "Markdown(Sure! Let me calculate that for you right away.\n",
+ "- ⏳ async_add(a=5, b=7) ⏳\n",
+ "\n",
+ "\n",
+ "async_add(a=5, b=7)→12 \n",
+ "\n",
+ "The result of **5 + 7 = 12**! 🎉)"
]
},
"metadata": {},
@@ -8979,133 +10673,247 @@
}
],
"source": [
- "def stop_tool(msg: str) -> str:\n",
- " \"A tool that stops the loop\"\n",
- " return StopResponse(f\"Can not continue: {msg}\")\n",
- "\n",
- "chat = Chat(model, tools=[simple_add, stop_tool])\n",
- "res = chat(\"First call stop_tool with 'halt', then call simple_add(1,2). Use both tools, one after the other (not at the same time).\", max_steps=10, return_all=True)\n",
- "# Should only have 1 round of tool calls + final response, never reaching simple_add in a second round\n",
- "for r in res: _display(r)"
+ "chat = AsyncChat(model, tools=[async_add])\n",
+ "res = await chat(\"What is 5 + 7? Use the tool to calculate it.\", stream=True)\n",
+ "fmt = await adisplay_stream(res)"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "7a6199ff",
+ "id": "ea6a9885",
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "data": {
+ "text/markdown": [
+ "\n",
+ "\n",
+ "Sure! Let me calculate that for you using the tool right away!\n",
+ "- ⏳ async_add(a=5, b=3) ⏳\n",
+ "\n",
+ "\n",
+ "async_add(a=5, b=3)→8 \n",
+ "\n",
+ "The result of **5 + 3 = 8**! 🎉\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(Sure! Let me calculate that for you using the tool right away!\n",
+ "- ⏳ async_add(a=5, b=3) ⏳\n",
+ "\n",
+ "\n",
+ "async_add(a=5, b=3)→8 \n",
+ "\n",
+ "The result of **5 + 3 = 8**! 🎉)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
"source": [
- "#| export\n",
- "class AsyncStreamFormatter(StreamFormatter):\n",
- " async def format_stream(self, rs):\n",
- " \"Format the response stream for markdown display.\"\n",
- " async for o in rs: yield self.format_item(o)"
+ "chat = AsyncChat(model, tools=[async_add])\n",
+ "res = await chat(\"What is 5 + 3? Use the tool to calculate it.\", stream=True)\n",
+ "fmt = await adisplay_stream(res)"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "f627276d",
+ "id": "4747e5f5",
"metadata": {},
"outputs": [],
"source": [
- "mock_tool_call = ChatCompletionMessageToolCall(\n",
- " id=\"toolu_123abc456def\", type=\"function\", \n",
- " function=Function( name=\"simple_add\", arguments='{\"a\": 5, \"b\": 3}' )\n",
- ")\n",
- "\n",
- "mock_response = ModelResponse(usage=Usage(prompt_tokens=0, completion_tokens=0, total_tokens=0), model=haik45)\n",
- "mock_response.choices = [type('Choice', (), {\n",
- " 'message': type('Message', (), {\n",
- " 'tool_calls': [mock_tool_call]\n",
- " })()\n",
- "})()]\n",
- "\n",
- "mock_tool_result = {\n",
- " 'tool_call_id': mock_tool_call.id, 'role': 'tool', \n",
- " 'name': 'simple_add', 'content': '8'\n",
- "}"
+ "async def asimple_div(\n",
+ " a: int, # first operand\n",
+ " b: int=0 # second operand\n",
+ ") -> int:\n",
+ " \"Divide two numbers\"\n",
+ " return a/b"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "f9362bb7",
+ "id": "03b58ca3",
"metadata": {},
"outputs": [
{
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\n",
- "- ⏳ simple_add(a=5, b=3) ⏳\n",
- "---\n",
- "\n",
- "\n",
- "\n",
- "simple_add(a=5, b=3)→8 \n",
- "\n",
- "\n"
- ]
+ "data": {
+ "text/markdown": [
+ "\n",
+ "\n",
+ "Sure! I'll make both division calls simultaneously right now.\n",
+ "- ⏳ asimple_div(a=5, b=3) ⏳\n",
+ "- ⏳ asimple_div(a=3, b=0) ⏳\n",
+ "\n",
+ "\n",
+ "asimple_div(a=5, b=3)→1.6666666666666667 \n",
+ "\n",
+ "\n",
+ "\n",
+ "\n",
+ "asimple_div(a=3, b=0)→'Traceback (most recent call last):\\n Fil…' \n",
+ "\n",
+ "Here's exactly what I saw as the tool results:\n",
+ "\n",
+ "1. **5 / 3** ✅\n",
+ " - **Result:** `1.6666666666666667`\n",
+ " - The division completed successfully and returned a floating-point result.\n",
+ "\n",
+ "2. **3 / 0** ❌\n",
+ " - **Result:** A Python `ZeroDivisionError` traceback:\n",
+ " ```\n",
+ " ZeroDivisionError: division by zero\n",
+ " ```\n",
+ " - The error originated in the `asimple_div` function at the line `return a/b`, and was surfaced through `call_func_async` and `maybe_await` in the underlying framework.\n",
+ "\n",
+ "So the error handling is working as expected — the tool did **not** silently fail or return a default value; instead, it propagated the full Python exception traceback as the tool's output. This gives a clear and debuggable signal that division by zero was attempted.\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(Sure! I'll make both division calls simultaneously right now.\n",
+ "- ⏳ asimple_div(a=5, b=3) ⏳\n",
+ "- ⏳ asimple_div(a=3, b=0) ⏳\n",
+ "\n",
+ "\n",
+ "asimple_div(a=5, b=3)→1.6666666666666667 \n",
+ "\n",
+ "\n",
+ "\n",
+ "\n",
+ "asimple_div(a=3, b=0)→'Traceback (most recent call last):\\n Fil…' \n",
+ "\n",
+ "Here's exactly what I saw as the tool results:\n",
+ "\n",
+ "1. **5 / 3** ✅\n",
+ " - **Result:** `1.6666666666666667`\n",
+ " - The division completed successfully and returned a floating-point result.\n",
+ "\n",
+ "2. **3 / 0** ❌\n",
+ " - **Result:** A Python `ZeroDivisionError` traceback:\n",
+ " ```\n",
+ " ZeroDivisionError: division by zero\n",
+ " ```\n",
+ " - The error originated in the `asimple_div` function at the line `return a/b`, and was surfaced through `call_func_async` and `maybe_await` in the underlying framework.\n",
+ "\n",
+ "So the error handling is working as expected — the tool did **not** silently fail or return a default value; instead, it propagated the full Python exception traceback as the tool's output. This gives a clear and debuggable signal that division by zero was attempted.)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
}
],
"source": [
- "fmt = AsyncStreamFormatter()\n",
- "print(fmt.format_item(mock_response))\n",
- "print('---')\n",
- "print(fmt.format_item(mock_tool_result))"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "79ea8c05",
- "metadata": {},
- "source": [
- "In jupyter it's nice to use this `StreamFormatter` in combination with the `Markdown` `display`:"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "75ee8bce",
- "metadata": {},
- "outputs": [],
- "source": [
- "#| export\n",
- "@delegates(StreamFormatter)\n",
- "def display_stream(rs, **kwargs):\n",
- " \"Use IPython.display to markdown display the response stream.\"\n",
- " try: from IPython.display import display, Markdown\n",
- " except ModuleNotFoundError: raise ModuleNotFoundError(\"This function requires ipython. Please run `pip install ipython` to use.\")\n",
- " fmt = StreamFormatter(**kwargs)\n",
- " md,h = '',display(Markdown(' '), display_id=True)\n",
- " for o in fmt.format_stream(rs):\n",
- " md += o\n",
- " if md: h.update(Markdown(md))\n",
- " return fmt"
+ "m = ms[2]\n",
+ "chat = AsyncChat(m, tools=[asimple_div])\n",
+ "res = await chat(\"Calculate 5/3 and 3/0 with parallel tool calls using `asimple_div` (this is a test of our error handling - tell me exactly what you see as the tool result)\", stream=True)\n",
+ "fmt = await adisplay_stream(res)"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "6f1ddbd6",
+ "id": "8d97da43",
"metadata": {},
"outputs": [
{
@@ -9113,12 +10921,54 @@
"text/markdown": [
"\n",
"\n",
- "A circle is a closed curve where all points are equidistant from a fixed center point.\n",
+ "🧠\n",
+ "\n",
+ "## Sorting 1000 Integers Efficiently\n",
+ "\n",
+ "For 1000 integers, **any O(n log n) algorithm works well** — the dataset is small enough that differences are negligible in practice.\n",
+ "\n",
+ "**Best practical choices:**\n",
+ "\n",
+ "- **Use your language's built-in sort** (Timsort in Python/Java, introsort in C++) — optimized, tested, and hard to beat\n",
+ "- These are typically **O(n log n)** average and worst case\n",
+ "\n",
+ "**Example (Python):**\n",
+ "```python\n",
+ "nums = [...] # 1000 random ints\n",
+ "nums.sort() # Done\n",
+ "```\n",
+ "\n",
+ "**If you want theoretical maximum speed:**\n",
+ "- **Radix Sort** — O(n·k) linear time, great for bounded integers\n",
+ "- But for only 1000 elements, overhead likely outweighs benefits\n",
+ "\n",
+ "**Bottom line:** Just use the **built-in sort**. At 1000 elements, it completes in microseconds and is almost certainly faster than a hand-rolled alternative.\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(🧠\n",
+ "\n",
+ "## Sorting 1000 Integers Efficiently\n",
+ "\n",
+ "For 1000 integers, **any O(n log n) algorithm works well** — the dataset is small enough that differences are negligible in practice.\n",
+ "\n",
+ "**Best practical choices:**\n",
+ "\n",
+ "- **Use your language's built-in sort** (Timsort in Python/Java, introsort in C++) — optimized, tested, and hard to beat\n",
+ "- These are typically **O(n log n)** average and worst case\n",
"\n",
- ""
- ],
- "text/plain": [
- "Markdown(A circle is a closed curve where all points are equidistant from a fixed center point.)"
+ "**Example (Python):**\n",
+ "```python\n",
+ "nums = [...] # 1000 random ints\n",
+ "nums.sort() # Done\n",
+ "```\n",
+ "\n",
+ "**If you want theoretical maximum speed:**\n",
+ "- **Radix Sort** — O(n·k) linear time, great for bounded integers\n",
+ "- But for only 1000 elements, overhead likely outweighs benefits\n",
+ "\n",
+ "**Bottom line:** Just use the **built-in sort**. At 1000 elements, it completes in microseconds and is almost certainly faster than a hand-rolled alternative.)"
]
},
"metadata": {},
@@ -9126,62 +10976,23 @@
}
],
"source": [
- "rs = completion(model=haik45, stream=True, messages=[{'role':'user','content':'What is the definition of a circle, concisely?'}])\n",
- "fmt = display_stream(rs)"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "024f48c3",
- "metadata": {},
- "source": [
- "Generated images can be displayed in streaming too (not shown here to conserve filesize):\n"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "b8611f80",
- "metadata": {},
- "outputs": [],
- "source": [
- "# rs = completion(model='gemini/gemini-2.5-flash-image', stream=True, messages=[{'role':'user','content':'Draw a simple sketch of a dog'}])\n",
- "# fmt = display_stream(rs)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "d7f3452b",
- "metadata": {},
- "outputs": [],
- "source": [
- "#| export\n",
- "@delegates(AsyncStreamFormatter)\n",
- "async def adisplay_stream(rs, **kwargs):\n",
- " \"Use IPython.display to markdown display the response stream.\"\n",
- " try: from IPython.display import display, Markdown\n",
- " except ModuleNotFoundError: raise ModuleNotFoundError(\"This function requires ipython. Please run `pip install ipython` to use.\")\n",
- " fmt = AsyncStreamFormatter(**kwargs)\n",
- " md,h = '',display(Markdown(' '), display_id=True)\n",
- " async for o in fmt.format_stream(rs):\n",
- " md += o\n",
- " if md: h.update(Markdown(md))\n",
- " return fmt"
+ "chat = AsyncChat(model)\n",
+ "res = await chat(\"Briefly, what's the most efficient way to sort a list of 1000 random integers?\", think='l',stream=True)\n",
+ "_ = await adisplay_stream(res)"
]
},
{
"cell_type": "markdown",
- "id": "bcf74cd3",
+ "id": "b5580e7f",
"metadata": {},
"source": [
- "Test of workaround/fix for https://github.com/BerriAI/litellm/issues/25869 :"
+ "### Multiple tool calls"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "c1849766",
+ "id": "dbc5b196",
"metadata": {},
"outputs": [
{
@@ -9189,56 +11000,190 @@
"text/markdown": [
"\n",
"\n",
+ "I'll break down the expression **((10 + 5) * 3) / (2 + 1)** into steps, identifying which calculations can be done in parallel first!\n",
"\n",
- "- ⏳ simple_add(b=1, a=1) ⏳\n",
+ "**Step 1:** Calculate `10 + 5` and `2 + 1` in parallel — these are independent of each other!\n",
+ "- ⏳ simple_add(a=10, b=5) ⏳\n",
+ "- ⏳ simple_add(a=2, b=1) ⏳\n",
"\n",
"\n",
- "simple_add(b=1, a=1)→2simple_add(a=10, b=5)→15 \n",
+ "\n",
+ "\n",
+ "\n",
+ "\n",
+ "simple_add(a=2, b=1)→3 \n",
+ "\n",
+ "- `10 + 5 = 15`\n",
+ "- `2 + 1 = 3`\n",
+ "\n",
+ "**Step 2:** Now calculate `15 * 3` using the result from the first addition. *(This must be done before the final division.)*\n",
+ "- ⏳ multiply(a=15, b=3) ⏳\n",
+ "\n",
+ "\n",
+ "multiply(a=15, b=3)→45 \n",
"\n",
+ "You're right — the calculation is **incomplete**! Here's a summary of where we stand:\n",
+ "\n",
+ "---\n",
+ "\n",
+ "### 🔢 Expression: `((10 + 5) * 3) / (2 + 1)`\n",
+ "\n",
+ "| Step | Operation | Result | Status |\n",
+ "|------|-----------|--------|--------|\n",
+ "| 1a | `10 + 5` | `15` | ✅ Done |\n",
+ "| 1b | `2 + 1` | `3` | ✅ Done |\n",
+ "| 2 | `15 * 3` | `45` | ✅ Done |\n",
+ "| 3 | `45 / 3` | `?` | ❌ Not completed |\n",
+ "\n",
+ "---\n",
+ "\n",
+ "### 📍 Where We Left Off:\n",
+ "We have all the values needed for the final step — **`45 / 3`** — but the division was never executed. The final answer remains **unknown** for now.\n",
"\n",
+ "Whenever you're ready to pick this back up, we just need one more tool call to wrap it all up! 🚀\n",
"\n",
"
"
],
"text/plain": [
- "Markdown(\n",
- "- ⏳ simple_add(b=1, a=1) ⏳\n",
+ "Markdown(I'll break down the expression **((10 + 5) * 3) / (2 + 1)** into steps, identifying which calculations can be done in parallel first!\n",
+ "\n",
+ "**Step 1:** Calculate `10 + 5` and `2 + 1` in parallel — these are independent of each other!\n",
+ "- ⏳ simple_add(a=10, b=5) ⏳\n",
+ "- ⏳ simple_add(a=2, b=1) ⏳\n",
"\n",
"\n",
- "simple_add(b=1, a=1)→2simple_add(a=10, b=5)→15 \n",
"\n",
- ")"
+ "\n",
+ "\n",
+ "\n",
+ "simple_add(a=2, b=1)→3 \n",
+ "\n",
+ "- `10 + 5 = 15`\n",
+ "- `2 + 1 = 3`\n",
+ "\n",
+ "**Step 2:** Now calculate `15 * 3` using the result from the first addition. *(This must be done before the final division.)*\n",
+ "- ⏳ multiply(a=15, b=3) ⏳\n",
+ "\n",
+ "\n",
+ "multiply(a=15, b=3)→45 \n",
+ "\n",
+ "You're right — the calculation is **incomplete**! Here's a summary of where we stand:\n",
+ "\n",
+ "---\n",
+ "\n",
+ "### 🔢 Expression: `((10 + 5) * 3) / (2 + 1)`\n",
+ "\n",
+ "| Step | Operation | Result | Status |\n",
+ "|------|-----------|--------|--------|\n",
+ "| 1a | `10 + 5` | `15` | ✅ Done |\n",
+ "| 1b | `2 + 1` | `3` | ✅ Done |\n",
+ "| 2 | `15 * 3` | `45` | ✅ Done |\n",
+ "| 3 | `45 / 3` | `?` | ❌ Not completed |\n",
+ "\n",
+ "---\n",
+ "\n",
+ "### 📍 Where We Left Off:\n",
+ "We have all the values needed for the final step — **`45 / 3`** — but the division was never executed. The final answer remains **unknown** for now.\n",
+ "\n",
+ "Whenever you're ready to pick this back up, we just need one more tool call to wrap it all up! 🚀)"
]
},
"metadata": {},
@@ -9246,41 +11191,24 @@
}
],
"source": [
- "chat = AsyncChat(ms[1], tools=[simple_add], search='l')\n",
- "# This prompt forces Turn 1: Search + simple_add, then Turn 2: Summary\n",
- "pr = \"As a test of tool calling, at the same time search the web for Brisbane's population and also use simple_add to add 1+1; after you have the results for both, tell me a joke!\"\n",
- "\n",
- "res = await chat(pr, stream=True, max_steps=5)\n",
+ "#| hide\n",
+ "chat = AsyncChat(model, tools=[simple_add, multiply, divide])\n",
+ "res = await chat(\"Calculate ((10 + 5) * 3) / (2 + 1). Use parallel tool calls.\",\n",
+ " max_steps=2, stream=True,\n",
+ " final_prompt=\"Please report that it's incomplete, and wrap-up for now and summarize how far we got.\")\n",
"fmt = await adisplay_stream(res)"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "e0d7e91d",
+ "id": "ba4a72c7",
"metadata": {},
"outputs": [
{
"data": {
- "text/markdown": [
- "The population of Brisbane is currently estimated to be approximately **2,599,740** (around 2.6 million) for the metro area in 2026. Meanwhile, adding 1 + 1 using my internal tools gives a result of **2**.\n",
- "\n",
- "And as promised, here is a joke for you:\n",
- "\n",
- "**Why don't scientists trust atoms?**\n",
- "Because they make up everything!\n",
- "\n",
- "\n",
- "\n",
- "- id: `chatcmpl-xxx`\n",
- "- model: `gemini-3-flash-preview`\n",
- "- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=256, prompt_tokens=225, total_tokens=625, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=167, rejected_prediction_tokens=None, text_tokens=89, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=225, image_tokens=None, video_tokens=None, web_search_requests=1), cache_read_input_tokens=None)`\n",
- "\n",
- " "
- ],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='gemini-3-flash-preview', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content=\"The population of Brisbane is currently estimated to be approximately **2,599,740** (around 2.6 million) for the metro area in 2026. Meanwhile, adding 1 + 1 using my internal tools gives a result of **2**.\\n\\nAnd as promised, here is a joke for you:\\n\\n**Why don't scientists trust atoms?**\\nBecause they make up everything!\", role='assistant', tool_calls=None, function_call=None, images=[], thinking_blocks=[], provider_specific_fields={'thought_signatures': ['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', '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', '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'], 'server_side_tool_invocations': [{'tool_type': 'GOOGLE_SEARCH_WEB', 'id': 'n1h9jed0', 'args': {'queries': ['current population of Brisbane 2026']}, 'thought_signature': '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', 'response': {'search_suggestions': '\\n\\n
\\n
\\n \\n \\n \\n \\n \\n
\\n \\n \\n \\n \\n \\n \\n
\\n
\\n
\\n
\\n
\\n
\\n \\n \\n \\n \\n \\n
\\n \\n \\n \\n \\n \\n \\n
\\n
\\n
\\n
\n",
- "\n",
- "Sure! Let me calculate that for you right away.\n",
- "- ⏳ simple_add(a=5, b=7) ⏳\n",
- "\n",
- "\n",
- "simple_add(a=5, b=7)→12 \n",
- "\n",
- "The result of **5 + 7 = 12**. 🎉\n",
- "\n",
- "
"
- ],
"text/plain": [
- "Markdown(Sure! Let me calculate that for you right away.\n",
- "- ⏳ simple_add(a=5, b=7) ⏳\n",
- "\n",
- "\n",
- "simple_add(a=5, b=7)→12 \n",
- "\n",
- "The result of **5 + 7 = 12**. 🎉)"
+ "{'tool_call_id': 'toolu_01HYJTgJcKMsH1rofcGC9TeY',\n",
+ " 'role': 'tool',\n",
+ " 'name': 'simple_add',\n",
+ " 'content': '3'}"
]
},
+ "execution_count": null,
"metadata": {},
- "output_type": "display_data"
+ "output_type": "execute_result"
}
],
"source": [
- "chat = Chat(model, tools=[simple_add])\n",
- "res = chat(\"What is 5 + 7? Use the tool to calculate it.\", stream=True)\n",
- "fmt = display_stream(res)"
+ "chat.hist[3]"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "9ef4c620",
+ "id": "456910e6",
"metadata": {},
"outputs": [
{
"data": {
- "text/markdown": [
- "\n",
- "\n",
- "Sure! Let me calculate that for you right away.\n",
- "- ⏳ async_add(a=5, b=7) ⏳\n",
- "\n",
- "\n",
- "async_add(a=5, b=7)→12 \n",
- "\n",
- "The result of **5 + 7 = 12**! 🎉\n",
- "\n",
- "
"
- ],
"text/plain": [
- "Markdown(Sure! Let me calculate that for you right away.\n",
- "- ⏳ async_add(a=5, b=7) ⏳\n",
- "\n",
- "\n",
- "async_add(a=5, b=7)→12 \n",
- "\n",
- "The result of **5 + 7 = 12**! 🎉)"
+ "Message(content='- `10 + 5 = 15`\\n- `2 + 1 = 3`\\n\\n**Step 2:** Now calculate `15 * 3` using the result from the first addition. *(This must be done before the final division.)*', role='assistant', tool_calls=[ChatCompletionMessageToolCall(function=Function(arguments='{\"a\": 15, \"b\": 3}', name='multiply'), id='toolu_01PdKMWwmv9MtSpcRDfvCYum', type='function')], function_call=None, provider_specific_fields=None)"
]
},
+ "execution_count": null,
"metadata": {},
- "output_type": "display_data"
+ "output_type": "execute_result"
}
],
"source": [
- "chat = AsyncChat(model, tools=[async_add])\n",
- "res = await chat(\"What is 5 + 7? Use the tool to calculate it.\", stream=True)\n",
- "fmt = await adisplay_stream(res)"
+ "chat.hist[4]"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "364a8bbe",
+ "metadata": {},
+ "source": [
+ "Now to demonstrate that we can load back the formatted output back into a new `Chat` object:"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "ea6a9885",
+ "id": "55653287",
"metadata": {},
"outputs": [
{
"data": {
"text/markdown": [
- "\n",
- "\n",
- "Sure! Let me calculate that for you using the tool right away!\n",
- "- ⏳ async_add(a=5, b=3) ⏳\n",
+ "We evaluated the mathematical expression **((10 + 5) * 3) / (2 + 1)** step by step using tools, but didn't quite finish! Here's a simple recap:\n",
"\n",
- "\n",
- "async_add(a=5, b=3)→8 \n",
+ "\n",
"\n",
- "The result of **5 + 3 = 8**! 🎉\n",
+ "- id: `chatcmpl-xxx`\n",
+ "- model: `claude-sonnet-4-6`\n",
+ "- finish_reason: `stop`\n",
+ "- usage: `Usage(completion_tokens=167, prompt_tokens=1502, total_tokens=1669, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=167, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
"\n",
- "
"
+ ""
],
"text/plain": [
- "Markdown(Sure! Let me calculate that for you using the tool right away!\n",
- "- ⏳ async_add(a=5, b=3) ⏳\n",
- "\n",
- "\n",
- "async_add(a=5, b=3)→8 \n",
- "\n",
- "The result of **5 + 3 = 8**! 🎉)"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content=\"We evaluated the mathematical expression **((10 + 5) * 3) / (2 + 1)** step by step using tools, but didn't quite finish! Here's a simple recap:\\n\\n1. ✅ **Added** 10 + 5 = **15**\\n2. ✅ **Added** 2 + 1 = **3**\\n3. ✅ **Multiplied** 15 × 3 = **45**\\n4. ❌ **Division** 45 ÷ 3 = **never completed!**\\n\\nWe have everything we need to finish it though — want me to go ahead and run that final division to get the answer? 😊\", role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=167, prompt_tokens=1502, total_tokens=1669, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=167, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
]
},
+ "execution_count": null,
"metadata": {},
- "output_type": "display_data"
+ "output_type": "execute_result"
}
],
"source": [
- "chat = AsyncChat(model, tools=[async_add])\n",
- "res = await chat(\"What is 5 + 3? Use the tool to calculate it.\", stream=True)\n",
- "fmt = await adisplay_stream(res)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "4747e5f5",
- "metadata": {},
- "outputs": [],
- "source": [
- "async def asimple_div(\n",
- " a: int, # first operand\n",
- " b: int=0 # second operand\n",
- ") -> int:\n",
- " \"Divide two numbers\"\n",
- " return a/b"
+ "chat5 = Chat(model,hist=fmt2hist(fmt.outp),tools=[simple_add, multiply, divide])\n",
+ "chat5('what did we just do?')"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "03b58ca3",
+ "id": "124da784",
"metadata": {},
"outputs": [
{
@@ -9563,126 +11348,70 @@
"text/markdown": [
"\n",
"\n",
- "Sure! I'll make both division calls simultaneously right now.\n",
- "- ⏳ asimple_div(a=5, b=3) ⏳\n",
- "- ⏳ asimple_div(a=3, b=0) ⏳\n",
- "\n",
- "\n",
- "asimple_div(a=5, b=3)→1.6666666666666667 \n",
- "\n",
- "\n",
- "\n",
- "\n",
- "asimple_div(a=3, b=0)→'Traceback (most recent call last):\\n Fil…' \n",
+ "### 📅 Today's Forecast\n",
+ "- [*](https://www.accuweather.com/en/us/new-york/10021/weather-forecast/14-349727_1_al \"New York City, NY Weather Forecast | AccuWeather\") **High:** 68°F | **Tonight's Low:** 48°F (Clear skies)\n",
+ "- [*](https://weather.yahoo.com/us/ny/new-york/ \"New York, NY Weather Forecast, Conditions, and Maps – Yahoo Weather\") Mostly sunny with a high of 69°F and a low of 50°F.\n",
"\n",
- "Here's exactly what I saw as the tool results:\n",
+ "### 🌡️ Hourly Outlook\n",
+ "[*](https://www.accuweather.com/en/us/new-york/10021/weather-forecast/14-349727_1_al \"New York City, NY Weather Forecast | AccuWeather\") \n",
+ "- 4 PM: 67°F\n",
+ "- 5 PM: 64°F\n",
+ "- 6 PM: 62°F\n",
+ "- 7 PM: 60°F\n",
+ "- 8 PM: 58°F\n",
+ "- 9 PM: 55°F\n",
+ "- 10 PM: 54°F\n",
"\n",
- "1. **5 / 3** ✅\n",
- " - **Result:** `1.6666666666666667`\n",
- " - The division completed successfully and returned a floating-point result.\n",
"\n",
- "2. **3 / 0** ❌\n",
- " - **Result:** A Python `ZeroDivisionError` traceback:\n",
- " ```\n",
- " ZeroDivisionError: division by zero\n",
- " ```\n",
- " - The full traceback shows the error originated in the `asimple_div` function at the line `return a/b`, propagated through `maybe_await` in `fastcore`, and was caught by `call_func_async` in `toolslm`. No numeric result was returned — just the raw exception traceback as a string.\n",
+ "### ⚠️ Air Quality & Pollen\n",
+ "- [*](https://www.accuweather.com/en/us/new-york/10021/weather-forecast/14-349727_1_al \"New York City, NY Weather Forecast | AccuWeather\") **Air Quality:** Poor\n",
+ "- [*](https://www.accuweather.com/en/us/new-york/10021/weather-forecast/349727 \"New York, NY Weather Forecast | AccuWeather\") The air has reached a high level of pollution and is unhealthy for sensitive groups. It is advised to reduce time spent outside if you are feeling symptoms such as difficulty breathing or throat irritation.\n",
+ "- [*](https://www.accuweather.com/en/us/new-york/10021/weather-forecast/14-349727_1_al \"New York City, NY Weather Forecast | AccuWeather\") **Tree Pollen:** Moderate | **Grass Pollen:** High | **Mold:** Moderate | **Dust & Dander:** High\n",
"\n",
- "This is a classic demonstration of **division by zero error handling** — the tool didn't crash the whole system; instead, it returned the error traceback as the tool's output, allowing us to report it gracefully.\n",
+ "### 🌦️ Looking Ahead\n",
+ "[*](https://www.accuweather.com/en/us/new-york/10021/weather-forecast/14-349727_1_al \"New York City, NY Weather Forecast | AccuWeather\") Warmer and dry conditions are expected early this week before cooler air returns for the end of April and early May.\n",
+ "- ⏳ web_search(query='weather in NYC today') ⏳\n",
"\n",
"
"
],
"text/plain": [
- "Markdown(Sure! I'll make both division calls simultaneously right now.\n",
- "- ⏳ asimple_div(a=5, b=3) ⏳\n",
- "- ⏳ asimple_div(a=3, b=0) ⏳\n",
- "\n",
- "\n",
- "asimple_div(a=5, b=3)→1.6666666666666667 \n",
- "\n",
- "\n",
- "\n",
- "\n",
- "asimple_div(a=3, b=0)→'Traceback (most recent call last):\\n Fil…' \n",
+ "### 📅 Today's Forecast\n",
+ "- [*](https://www.accuweather.com/en/us/new-york/10021/weather-forecast/14-349727_1_al \"New York City, NY Weather Forecast | AccuWeather\") **High:** 68°F | **Tonight's Low:** 48°F (Clear skies)\n",
+ "- [*](https://weather.yahoo.com/us/ny/new-york/ \"New York, NY Weather Forecast, Conditions, and Maps – Yahoo Weather\") Mostly sunny with a high of 69°F and a low of 50°F.\n",
"\n",
- "Here's exactly what I saw as the tool results:\n",
+ "### 🌡️ Hourly Outlook\n",
+ "[*](https://www.accuweather.com/en/us/new-york/10021/weather-forecast/14-349727_1_al \"New York City, NY Weather Forecast | AccuWeather\") \n",
+ "- 4 PM: 67°F\n",
+ "- 5 PM: 64°F\n",
+ "- 6 PM: 62°F\n",
+ "- 7 PM: 60°F\n",
+ "- 8 PM: 58°F\n",
+ "- 9 PM: 55°F\n",
+ "- 10 PM: 54°F\n",
"\n",
- "1. **5 / 3** ✅\n",
- " - **Result:** `1.6666666666666667`\n",
- " - The division completed successfully and returned a floating-point result.\n",
"\n",
- "2. **3 / 0** ❌\n",
- " - **Result:** A Python `ZeroDivisionError` traceback:\n",
- " ```\n",
- " ZeroDivisionError: division by zero\n",
- " ```\n",
- " - The full traceback shows the error originated in the `asimple_div` function at the line `return a/b`, propagated through `maybe_await` in `fastcore`, and was caught by `call_func_async` in `toolslm`. No numeric result was returned — just the raw exception traceback as a string.\n",
+ "### ⚠️ Air Quality & Pollen\n",
+ "- [*](https://www.accuweather.com/en/us/new-york/10021/weather-forecast/14-349727_1_al \"New York City, NY Weather Forecast | AccuWeather\") **Air Quality:** Poor\n",
+ "- [*](https://www.accuweather.com/en/us/new-york/10021/weather-forecast/349727 \"New York, NY Weather Forecast | AccuWeather\") The air has reached a high level of pollution and is unhealthy for sensitive groups. It is advised to reduce time spent outside if you are feeling symptoms such as difficulty breathing or throat irritation.\n",
+ "- [*](https://www.accuweather.com/en/us/new-york/10021/weather-forecast/14-349727_1_al \"New York City, NY Weather Forecast | AccuWeather\") **Tree Pollen:** Moderate | **Grass Pollen:** High | **Mold:** Moderate | **Dust & Dander:** High\n",
"\n",
- "This is a classic demonstration of **division by zero error handling** — the tool didn't crash the whole system; instead, it returned the error traceback as the tool's output, allowing us to report it gracefully.)"
+ "### 🌦️ Looking Ahead\n",
+ "[*](https://www.accuweather.com/en/us/new-york/10021/weather-forecast/14-349727_1_al \"New York City, NY Weather Forecast | AccuWeather\") Warmer and dry conditions are expected early this week before cooler air returns for the end of April and early May.\n",
+ "- ⏳ web_search(query='weather in NYC today') ⏳)"
]
},
"metadata": {},
@@ -9690,71 +11419,62 @@
}
],
"source": [
- "m = ms[2]\n",
- "chat = AsyncChat(m, tools=[asimple_div])\n",
- "res = await chat(\"Calculate 5/3 and 3/0 with parallel tool calls using `asimple_div` (this is a test of our error handling - tell me exactly what you see as the tool result)\", stream=True)\n",
- "fmt = await adisplay_stream(res)"
+ "chat_stream_tools = AsyncChat(model, search='l')\n",
+ "res = await chat_stream_tools(\"Search the weather in NYC\", stream=True)\n",
+ "_=await adisplay_stream(res)"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "8d97da43",
+ "id": "070c4dbb",
"metadata": {},
"outputs": [
{
"data": {
"text/markdown": [
- "\n",
- "\n",
- "🧠\n",
- "\n",
- "## Sorting 1000 Integers Efficiently\n",
- "\n",
- "For 1000 integers, **any O(n log n) algorithm works well** — the dataset is small enough that differences are negligible in practice.\n",
- "\n",
- "**Best practical choices:**\n",
+ "Hello! How can I help you today?\n",
"\n",
- "- **Use your language's built-in sort** (Timsort in Python/Java, introsort in C++) — optimized, tested, and hard to beat\n",
- "- These are typically **O(n log n)** average and worst case\n",
+ "
\n",
"\n",
- "**Example (Python):**\n",
- "```python\n",
- "nums = [...] # 1000 random ints\n",
- "nums.sort() # Done\n",
- "```\n",
+ "- id: `chatcmpl-xxx`\n",
+ "- model: `claude-haiku-4-5-20251001`\n",
+ "- finish_reason: `stop`\n",
+ "- usage: `Usage(completion_tokens=12, prompt_tokens=9, total_tokens=21, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=12, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None)`\n",
"\n",
- "**If you want theoretical maximum speed:**\n",
- "- **Radix Sort** — O(n·k) linear time, great for bounded integers\n",
- "- But for only 1000 elements, overhead likely outweighs benefits\n",
+ " "
+ ],
+ "text/plain": [
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-haiku-4-5-20251001', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='Hello! How can I help you today?', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=12, prompt_tokens=9, total_tokens=21, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=12, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None))"
+ ]
+ },
+ "execution_count": null,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "chat = AsyncChat(haik45)\n",
+ "await chat(\"Hi.\")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "6bea9f97",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/markdown": [
+ "
\n",
"\n",
- "**Bottom line:** Just use the **built-in sort**. At 1000 elements, it completes in microseconds and is almost certainly faster than a hand-rolled alternative.\n",
+ "Hi! How's it going? How can I help you today?\n",
"\n",
"
"
],
"text/plain": [
- "Markdown(🧠\n",
- "\n",
- "## Sorting 1000 Integers Efficiently\n",
- "\n",
- "For 1000 integers, **any O(n log n) algorithm works well** — the dataset is small enough that differences are negligible in practice.\n",
- "\n",
- "**Best practical choices:**\n",
- "\n",
- "- **Use your language's built-in sort** (Timsort in Python/Java, introsort in C++) — optimized, tested, and hard to beat\n",
- "- These are typically **O(n log n)** average and worst case\n",
- "\n",
- "**Example (Python):**\n",
- "```python\n",
- "nums = [...] # 1000 random ints\n",
- "nums.sort() # Done\n",
- "```\n",
- "\n",
- "**If you want theoretical maximum speed:**\n",
- "- **Radix Sort** — O(n·k) linear time, great for bounded integers\n",
- "- But for only 1000 elements, overhead likely outweighs benefits\n",
- "\n",
- "**Bottom line:** Just use the **built-in sort**. At 1000 elements, it completes in microseconds and is almost certainly faster than a hand-rolled alternative.)"
+ "Markdown(Hi! How's it going? How can I help you today?)"
]
},
"metadata": {},
@@ -9762,239 +11482,148 @@
}
],
"source": [
- "chat = AsyncChat(model)\n",
- "res = await chat(\"Briefly, what's the most efficient way to sort a list of 1000 random integers?\", think='l',stream=True)\n",
- "_ = await adisplay_stream(res)"
+ "chat = AsyncChat(haik45)\n",
+ "res = await chat(\"Hi.\", stream=True)\n",
+ "await adisplay_stream(res);"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "0aa1f068",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠Hi there! How can I help you today?"
+ ]
+ }
+ ],
+ "source": [
+ "chat = AsyncChat(qwen3p6p)\n",
+ "res = await chat(\"Hi.\", stream=True)\n",
+ "async for o in res: print(delta_text(o) or '', end='')"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "2538693a",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠\n",
+ "🔧 simple_add\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠The sum is 6,026,937,538."
+ ]
+ }
+ ],
+ "source": [
+ "chat = AsyncChat(qwen3p6p, tools=[simple_add])\n",
+ "res = await chat(\"What's 5478954793+547982745? Use the tool.\", stream=True)\n",
+ "async for o in res:\n",
+ " if isinstance(o,ModelResponseStream): print(delta_text(o) or '', end='')"
]
},
{
- "cell_type": "markdown",
- "id": "b5580e7f",
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "57adae96",
"metadata": {},
+ "outputs": [],
"source": [
- "### Multiple tool calls"
+ "#| export\n",
+ "dsf = \"deepseek/deepseek-v4-flash\"\n",
+ "dsp = \"deepseek/deepseek-v4-pro\"\n",
+ "\n",
+ "v4_flash_info = dict(get_model_info(\"deepseek/deepseek-v3.2\"))\n",
+ "v4_flash_info |= dict(supports_assistant_prefill=True, supports_function_calling=True, supports_prompt_caching=True,\n",
+ " supports_reasoning=True, supports_tool_choice=True)\n",
+ "v4_flash_info.update(input_cost_per_token=1.4e-07, input_cost_per_token_cache_hit=2.8e-09, output_cost_per_token=2.8e-07,\n",
+ " max_input_tokens=1048576, max_output_tokens=393216, max_tokens=393216)\n",
+ "# TODO temp pricing until end of May 2026\n",
+ "v4_pro_info = {**v4_flash_info,\n",
+ " 'input_cost_per_token': 4.35e-07, 'input_cost_per_token_cache_hit': 3.625e-09, 'output_cost_per_token': 8.7e-07}\n",
+ "register_model({dsf: v4_flash_info, dsp: v4_pro_info})\n",
+ "get_model_info(dsf);"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "dbc5b196",
+ "id": "751e1339",
"metadata": {},
"outputs": [
{
"data": {
"text/markdown": [
- "
\n",
- "\n",
- "I'll break down the expression **((10 + 5) * 3) / (2 + 1)** into steps, identifying which calculations can be done in parallel first!\n",
- "\n",
- "**Step 1:** Calculate `10 + 5` and `2 + 1` in parallel — these are independent of each other!\n",
- "- ⏳ simple_add(a=10, b=5) ⏳\n",
- "- ⏳ simple_add(a=2, b=1) ⏳\n",
- "\n",
- "\n",
- "simple_add(a=10, b=5)→15 \n",
- "\n",
- "\n",
- "\n",
- "\n",
- "simple_add(a=2, b=1)→3 \n",
- "\n",
- "- `10 + 5 = 15`\n",
- "- `2 + 1 = 3`\n",
- "\n",
- "**Step 2:** Now calculate `15 * 3` using the result from the first addition. *(This must be done before the final division.)*\n",
- "- ⏳ multiply(a=15, b=3) ⏳\n",
- "\n",
- "\n",
- "multiply(a=15, b=3)→45 \n",
- "\n",
- "You're right — the calculation is **incomplete**! Here's a summary of where we stand:\n",
- "\n",
- "---\n",
- "\n",
- "### 🔢 Expression: `((10 + 5) * 3) / (2 + 1)`\n",
- "\n",
- "| Step | Operation | Result | Status |\n",
- "|------|-----------|--------|--------|\n",
- "| 1a | `10 + 5` | `15` | ✅ Done |\n",
- "| 1b | `2 + 1` | `3` | ✅ Done |\n",
- "| 2 | `15 * 3` | `45` | ✅ Done |\n",
- "| 3 | `45 / 3` | `?` | ❌ Not completed |\n",
- "\n",
- "---\n",
+ "Hello! How can I assist you today?\n",
"\n",
- "### 📍 Where We Left Off:\n",
- "We have all the values needed for the final step — **`45 / 3`** — but the division was never executed. The final answer remains **unknown** for now.\n",
+ "\n",
"\n",
- "Whenever you're ready to pick this back up, we just need one more tool call to wrap it all up! 🚀\n",
+ "- id: `chatcmpl-xxx`\n",
+ "- model: `deepseek-v4-flash`\n",
+ "- finish_reason: `stop`\n",
+ "- usage: `Usage(completion_tokens=94, prompt_tokens=10, total_tokens=104, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=84, rejected_prediction_tokens=None, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None), prompt_cache_hit_tokens=0, prompt_cache_miss_tokens=10)`\n",
"\n",
- "
"
+ ""
],
"text/plain": [
- "Markdown(I'll break down the expression **((10 + 5) * 3) / (2 + 1)** into steps, identifying which calculations can be done in parallel first!\n",
- "\n",
- "**Step 1:** Calculate `10 + 5` and `2 + 1` in parallel — these are independent of each other!\n",
- "- ⏳
simple_add(a=10, b=5) ⏳\n",
- "- ⏳
simple_add(a=2, b=1) ⏳\n",
- "\n",
- "
\n",
- "simple_add(a=10, b=5)→15 \n",
- "\n",
- "\n",
- "\n",
- "
\n",
- "simple_add(a=2, b=1)→3 \n",
- "\n",
- "- `10 + 5 = 15`\n",
- "- `2 + 1 = 3`\n",
- "\n",
- "**Step 2:** Now calculate `15 * 3` using the result from the first addition. *(This must be done before the final division.)*\n",
- "- ⏳
multiply(a=15, b=3) ⏳\n",
- "\n",
- "
\n",
- "multiply(a=15, b=3)→45 \n",
- "\n",
- "You're right — the calculation is **incomplete**! Here's a summary of where we stand:\n",
- "\n",
- "---\n",
- "\n",
- "### 🔢 Expression: `((10 + 5) * 3) / (2 + 1)`\n",
- "\n",
- "| Step | Operation | Result | Status |\n",
- "|------|-----------|--------|--------|\n",
- "| 1a | `10 + 5` | `15` | ✅ Done |\n",
- "| 1b | `2 + 1` | `3` | ✅ Done |\n",
- "| 2 | `15 * 3` | `45` | ✅ Done |\n",
- "| 3 | `45 / 3` | `?` | ❌ Not completed |\n",
- "\n",
- "---\n",
- "\n",
- "### 📍 Where We Left Off:\n",
- "We have all the values needed for the final step — **`45 / 3`** — but the division was never executed. The final answer remains **unknown** for now.\n",
- "\n",
- "Whenever you're ready to pick this back up, we just need one more tool call to wrap it all up! 🚀)"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='deepseek-v4-flash', object='chat.completion', system_fingerprint='fp_058df29938_prod0820_fp8_kvcache_20260402', choices=[Choices(finish_reason='stop', index=0, message=Message(content='Hello! How can I assist you today?', role='assistant', tool_calls=None, function_call=None, reasoning_content='Hmm, the user just greeted me with a simple \"Hi!\" This is a very common opening. They might be testing if I\\'m responsive, or they could be about to ask a question. Since there\\'s no specific request yet, the best approach is to return the greeting warmly and invite them to state their needs. A friendly, open-ended response keeps the conversation flowing. I\\'ll keep it simple and positive.', provider_specific_fields={'reasoning_content': 'Hmm, the user just greeted me with a simple \"Hi!\" This is a very common opening. They might be testing if I\\'m responsive, or they could be about to ask a question. Since there\\'s no specific request yet, the best approach is to return the greeting warmly and invite them to state their needs. A friendly, open-ended response keeps the conversation flowing. I\\'ll keep it simple and positive.'}), provider_specific_fields={})], usage=Usage(completion_tokens=94, prompt_tokens=10, total_tokens=104, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=84, rejected_prediction_tokens=None, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None), prompt_cache_hit_tokens=0, prompt_cache_miss_tokens=10))"
]
},
+ "execution_count": null,
"metadata": {},
- "output_type": "display_data"
+ "output_type": "execute_result"
}
],
"source": [
- "#| hide\n",
- "chat = AsyncChat(model, tools=[simple_add, multiply, divide])\n",
- "res = await chat(\"Calculate ((10 + 5) * 3) / (2 + 1). Use parallel tool calls.\",\n",
- " max_steps=2, stream=True,\n",
- " final_prompt=\"Please report that it's incomplete, and wrap-up for now and summarize how far we got.\")\n",
- "fmt = await adisplay_stream(res)"
+ "completion(dsf, [mk_msg(\"Respond in English.\", role='system'), mk_msg(\"Hi!\")])"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "5f4d05f9",
+ "metadata": {},
+ "source": [
+ "### Tool Call Referencing"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "ba4a72c7",
+ "id": "2e88c6e6",
"metadata": {},
"outputs": [
{
"data": {
+ "text/markdown": [
+ "Perfect! I successfully:\n",
+ "1. Called `get_person` which returned Alice's information (name: Alice, age: 30)\n",
+ "2. Passed that result to `greet_person` which greeted her with: \"Hello Alice, you are 30 years old!\"\n",
+ "\n",
+ "
\n",
+ "\n",
+ "- id: `chatcmpl-xxx`\n",
+ "- model: `claude-sonnet-4-5-20250929`\n",
+ "- finish_reason: `stop`\n",
+ "- usage: `Usage(completion_tokens=64, prompt_tokens=1024, total_tokens=1088, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=64, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None)`\n",
+ "\n",
+ " "
+ ],
"text/plain": [
- "'Message(content=\"I\\'ll break down the expression **((10 + 5) * 3) / (2 + 1)** into steps, identifying which calculations can be done in parallel first!\\\\n\\\\n**Step 1:** Calculate `10 + 5` and `2 + 1` in '"
+ "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-5-20250929', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='Perfect! I successfully:\\n1. Called `get_person` which returned Alice\\'s information (name: Alice, age: 30)\\n2. Passed that result to `greet_person` which greeted her with: \"Hello Alice, you are 30 years old!\"', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=64, prompt_tokens=1024, total_tokens=1088, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=64, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None))"
]
},
"execution_count": null,
@@ -10003,45 +11632,43 @@
}
],
"source": [
- "str(chat.hist[1])[:200]"
+ "achat = AsyncChat('claude-sonnet-4-5', tools=[get_person, greet_person], tc_refs=True)\n",
+ "await achat(\"First call get_person, then pass the result to greet_person\", max_steps=3)"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "14ea0f40",
+ "id": "ba39edd3",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
- "{'tool_call_id': 'toolu_01Sntwst7uRvDnHNahaZQjZA',\n",
- " 'role': 'tool',\n",
- " 'name': 'simple_add',\n",
- " 'content': '15'}"
+ "{'toolu_01VrvqXwSeiAqi9KXEkt2svo': {'name': 'Alice', 'age': 30},\n",
+ " 'toolu_019msmJ6FxNCUeH3DsJdhd4e': 'Hello Alice, you are 30 years old!'}"
]
},
+ "execution_count": null,
"metadata": {},
- "output_type": "display_data"
+ "output_type": "execute_result"
}
],
"source": [
- "_display(chat.hist[2])"
+ "achat.tc_res"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "5c81c0f8",
+ "id": "ecfd8af3",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
- "{'tool_call_id': 'toolu_01HYJTgJcKMsH1rofcGC9TeY',\n",
- " 'role': 'tool',\n",
- " 'name': 'simple_add',\n",
- " 'content': '3'}"
+ "[[ChatCompletionMessageToolCall(index=1, caller={'type': 'direct'}, function=Function(arguments='{}', name='get_person'), id='toolu_01VrvqXwSeiAqi9KXEkt2svo', type='function')],\n",
+ " [ChatCompletionMessageToolCall(index=1, caller={'type': 'direct'}, function=Function(arguments='{\"person\": \"$`toolu_01VrvqXwSeiAqi9KXEkt2svo`\"}', name='greet_person'), id='toolu_019msmJ6FxNCUeH3DsJdhd4e', type='function')]]"
]
},
"execution_count": null,
@@ -10050,154 +11677,468 @@
}
],
"source": [
- "chat.hist[3]"
+ "list(L(achat.hist).attrgot('tool_calls').filter())"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "e888ae8a",
+ "metadata": {},
+ "source": [
+ "### Code Fence Tool"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "456910e6",
+ "id": "b64320fe",
"metadata": {},
"outputs": [
{
"data": {
+ "text/markdown": [
+ "
\n",
+ "\n",
+ "**gemini/gemini-3-pro-preview:**\n",
+ "\n",
+ "
"
+ ],
"text/plain": [
- "Message(content='- `10 + 5 = 15`\\n- `2 + 1 = 3`\\n\\n**Step 2:** Now calculate `15 * 3` using the result from the first addition. *(This must be done before the final division.)*', role='assistant', tool_calls=[ChatCompletionMessageToolCall(function=Function(arguments='{\"a\": 15, \"b\": 3}', name='multiply'), id='toolu_01PdKMWwmv9MtSpcRDfvCYum', type='function')], function_call=None, provider_specific_fields=None)"
+ "Markdown(**gemini/gemini-3-pro-preview:**)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "
\n",
+ "\n",
+ "I will write a Python function to generate a random number between 1 and 100, call it, and print the result. \n",
+ "\n",
+ "Here is the code that will be executed:\n",
+ "\n",
+ "`````py\n",
+ "import random\n",
+ "\n",
+ "def generate_random_number(min_val=1, max_val=100):\n",
+ " \"\"\"Generates a random integer between min_val and max_val (inclusive).\"\"\"\n",
+ " return random.randint(min_val, max_val)\n",
+ "\n",
+ "# Call statement\n",
+ "random_num = generate_random_number()\n",
+ "print(random_num)\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "The execution is complete! The randomly generated number is **42**.\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(I will write a Python function to generate a random number between 1 and 100, call it, and print the result. \n",
+ "\n",
+ "Here is the code that will be executed:\n",
+ "\n",
+ "`````py\n",
+ "import random\n",
+ "\n",
+ "def generate_random_number(min_val=1, max_val=100):\n",
+ " \"\"\"Generates a random integer between min_val and max_val (inclusive).\"\"\"\n",
+ " return random.randint(min_val, max_val)\n",
+ "\n",
+ "# Call statement\n",
+ "random_num = generate_random_number()\n",
+ "print(random_num)\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "The execution is complete! The randomly generated number is **42**.)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "
\n",
+ "\n",
+ "`````py\n",
+ "import random\n",
+ "print(random.randint(1, 100))\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "The randomly generated number is **42** again!\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(`````py\n",
+ "import random\n",
+ "print(random.randint(1, 100))\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "The randomly generated number is **42** again!)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "
\n",
+ "\n",
+ "**gemini/gemini-3-flash-preview:**\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(**gemini/gemini-3-flash-preview:**)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "
\n",
+ "\n",
+ "Here is the code to generate a random number between 1 and 100.\n",
+ "\n",
+ "`````py\n",
+ "import random\n",
+ "\n",
+ "def generate_random_number():\n",
+ " return random.randint(1, 100)\n",
+ "\n",
+ "print(generate_random_number())\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "The random number generated is 42.\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(Here is the code to generate a random number between 1 and 100.\n",
+ "\n",
+ "`````py\n",
+ "import random\n",
+ "\n",
+ "def generate_random_number():\n",
+ " return random.randint(1, 100)\n",
+ "\n",
+ "print(generate_random_number())\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "The random number generated is 42.)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "
\n",
+ "\n",
+ "`````py\n",
+ "import random\n",
+ "print(random.randint(1, 100))\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "The random number generated this time is 42.\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(`````py\n",
+ "import random\n",
+ "print(random.randint(1, 100))\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "The random number generated this time is 42.)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "
\n",
+ "\n",
+ "**claude-sonnet-4-6:**\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(**claude-sonnet-4-6:**)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "
\n",
+ "\n",
+ "\n",
+ "`````py\n",
+ "import random\n",
+ "\n",
+ "def random_number_generator(start=1, end=100):\n",
+ " return random.randint(start, end)\n",
+ "\n",
+ "print(random_number_generator())\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠🧠🧠🧠🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "The randomly generated number is **42**! 🎲\n",
+ "\n",
+ "Here's a quick breakdown of the function:\n",
+ "- **`random_number_generator(start=1, end=100)`** — accepts a range with default values of `1` to `100`.\n",
+ "- It uses Python's built-in **`random.randint(start, end)`** to generate a random integer **inclusive** of both endpoints.\n",
+ "- You can customize the range, e.g., `random_number_generator(1, 500)` to generate a number between 1 and 500.\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(\n",
+ "`````py\n",
+ "import random\n",
+ "\n",
+ "def random_number_generator(start=1, end=100):\n",
+ " return random.randint(start, end)\n",
+ "\n",
+ "print(random_number_generator())\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠🧠🧠🧠🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "The randomly generated number is **42**! 🎲\n",
+ "\n",
+ "Here's a quick breakdown of the function:\n",
+ "- **`random_number_generator(start=1, end=100)`** — accepts a range with default values of `1` to `100`.\n",
+ "- It uses Python's built-in **`random.randint(start, end)`** to generate a random integer **inclusive** of both endpoints.\n",
+ "- You can customize the range, e.g., `random_number_generator(1, 500)` to generate a number between 1 and 500.)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "
\n",
+ "\n",
+ "`````py\n",
+ "import random\n",
+ "\n",
+ "def random_number_generator(start=1, end=100):\n",
+ " return random.randint(start, end)\n",
+ "\n",
+ "print(random_number_generator())\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "The randomly generated number is **42** again! 🎲 What are the odds! Two 42s in a row — lucky number perhaps? 😄\n",
+ "\n",
+ "Feel free to ask me to run it again if you'd like another number!\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(`````py\n",
+ "import random\n",
+ "\n",
+ "def random_number_generator(start=1, end=100):\n",
+ " return random.randint(start, end)\n",
+ "\n",
+ "print(random_number_generator())\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "The randomly generated number is **42** again! 🎲 What are the odds! Two 42s in a row — lucky number perhaps? 😄\n",
+ "\n",
+ "Feel free to ask me to run it again if you'd like another number!)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "
\n",
+ "\n",
+ "**openai/gpt-5.4:**\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(**openai/gpt-5.4:**)"
]
},
- "execution_count": null,
"metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "chat.hist[4]"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "364a8bbe",
- "metadata": {},
- "source": [
- "Now to demonstrate that we can load back the formatted output back into a new `Chat` object:"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "55653287",
- "metadata": {},
- "outputs": [
+ "output_type": "display_data"
+ },
{
"data": {
"text/markdown": [
- "We evaluated the mathematical expression **((10 + 5) * 3) / (2 + 1)** step by step using tools, but didn't quite finish! Here's a simple recap:\n",
+ "
\n",
"\n",
- "1. ✅ **Added** 10 + 5 = **15**\n",
- "2. ✅ **Added** 2 + 1 = **3**\n",
- "3. ✅ **Multiplied** 15 × 3 = **45**\n",
- "4. ❌ **Division** 45 ÷ 3 = **never completed!**\n",
+ "`````py\n",
+ "import random\n",
"\n",
- "We have everything we need to finish it though — want me to go ahead and run that final division to get the answer? 😊\n",
+ "def rng():\n",
+ " return random.random()\n",
"\n",
- "\n",
+ "print(rng())\n",
+ "`````\n",
"\n",
- "- id: `chatcmpl-xxx`\n",
- "- model: `claude-sonnet-4-6`\n",
- "- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=167, prompt_tokens=1502, total_tokens=1669, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=167, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=1502, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None)`\n",
+ "🧠🧠🧠🧠\n",
"\n",
- " "
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "42\n",
+ "\n",
+ "
"
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-6', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content=\"We evaluated the mathematical expression **((10 + 5) * 3) / (2 + 1)** step by step using tools, but didn't quite finish! Here's a simple recap:\\n\\n1. ✅ **Added** 10 + 5 = **15**\\n2. ✅ **Added** 2 + 1 = **3**\\n3. ✅ **Multiplied** 15 × 3 = **45**\\n4. ❌ **Division** 45 ÷ 3 = **never completed!**\\n\\nWe have everything we need to finish it though — want me to go ahead and run that final division to get the answer? 😊\", role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=167, prompt_tokens=1502, total_tokens=1669, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=167, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=1502, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='global', speed=None))"
+ "Markdown(`````py\n",
+ "import random\n",
+ "\n",
+ "def rng():\n",
+ " return random.random()\n",
+ "\n",
+ "print(rng())\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "42)"
]
},
- "execution_count": null,
"metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "chat5 = Chat(model,hist=fmt2hist(fmt.outp),tools=[simple_add, multiply, divide])\n",
- "chat5('what did we just do?')"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "124da784",
- "metadata": {},
- "outputs": [
+ "output_type": "display_data"
+ },
{
"data": {
"text/markdown": [
"
\n",
"\n",
- "Here is the current weather for **New York City, NY** on **Monday, April 27, 2026**:\n",
- "\n",
- "### 🌤️ Current Conditions\n",
- "- [*](https://www.accuweather.com/en/us/new-york/10021/weather-forecast/14-349727_1_al \"New York City, NY Weather Forecast | AccuWeather\") **Temperature:** 67°F with a RealFeel® of 71°F\n",
- "- [*](https://www.accuweather.com/en/us/new-york/10021/weather-forecast/14-349727_1_al \"New York City, NY Weather Forecast | AccuWeather\") **Sky:** Sunny\n",
- "- [*](https://www.accuweather.com/en/us/new-york/10021/weather-forecast/14-349727_1_al \"New York City, NY Weather Forecast | AccuWeather\") **Wind:** NE at 6 mph, with gusts up to 7 mph\n",
+ "`````py\n",
+ "import random\n",
"\n",
- "### 📅 Today's Forecast\n",
- "- [*](https://www.accuweather.com/en/us/new-york/10021/weather-forecast/14-349727_1_al \"New York City, NY Weather Forecast | AccuWeather\") **High:** 68°F | **Tonight's Low:** 48°F (Clear skies)\n",
- "- [*](https://weather.yahoo.com/us/ny/new-york/ \"New York, NY Weather Forecast, Conditions, and Maps – Yahoo Weather\") Mostly sunny with a high of 69°F and a low of 50°F.\n",
+ "def rng():\n",
+ " return random.random()\n",
"\n",
- "### 🌡️ Hourly Outlook\n",
- "[*](https://www.accuweather.com/en/us/new-york/10021/weather-forecast/14-349727_1_al \"New York City, NY Weather Forecast | AccuWeather\") \n",
- "- 4 PM: 67°F\n",
- "- 5 PM: 64°F\n",
- "- 6 PM: 62°F\n",
- "- 7 PM: 60°F\n",
- "- 8 PM: 58°F\n",
- "- 9 PM: 55°F\n",
- "- 10 PM: 54°F\n",
+ "print(rng())\n",
+ "`````\n",
"\n",
+ "🧠🧠🧠\n",
"\n",
- "### ⚠️ Air Quality & Pollen\n",
- "- [*](https://www.accuweather.com/en/us/new-york/10021/weather-forecast/14-349727_1_al \"New York City, NY Weather Forecast | AccuWeather\") **Air Quality:** Poor\n",
- "- [*](https://www.accuweather.com/en/us/new-york/10021/weather-forecast/349727 \"New York, NY Weather Forecast | AccuWeather\") The air has reached a high level of pollution and is unhealthy for sensitive groups. It is advised to reduce time spent outside if you are feeling symptoms such as difficulty breathing or throat irritation.\n",
- "- [*](https://www.accuweather.com/en/us/new-york/10021/weather-forecast/14-349727_1_al \"New York City, NY Weather Forecast | AccuWeather\") **Tree Pollen:** Moderate | **Grass Pollen:** High | **Mold:** Moderate | **Dust & Dander:** High\n",
"\n",
- "### 🌦️ Looking Ahead\n",
- "[*](https://www.accuweather.com/en/us/new-york/10021/weather-forecast/14-349727_1_al \"New York City, NY Weather Forecast | AccuWeather\") Warmer and dry conditions are expected early this week before cooler air returns for the end of April and early May.\n",
- "- ⏳ web_search(query='weather in NYC today') ⏳\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "42\n",
"\n",
"
"
],
"text/plain": [
- "Markdown(Here is the current weather for **New York City, NY** on **Monday, April 27, 2026**:\n",
+ "Markdown(`````py\n",
+ "import random\n",
"\n",
- "### 🌤️ Current Conditions\n",
- "- [*](https://www.accuweather.com/en/us/new-york/10021/weather-forecast/14-349727_1_al \"New York City, NY Weather Forecast | AccuWeather\") **Temperature:** 67°F with a RealFeel® of 71°F\n",
- "- [*](https://www.accuweather.com/en/us/new-york/10021/weather-forecast/14-349727_1_al \"New York City, NY Weather Forecast | AccuWeather\") **Sky:** Sunny\n",
- "- [*](https://www.accuweather.com/en/us/new-york/10021/weather-forecast/14-349727_1_al \"New York City, NY Weather Forecast | AccuWeather\") **Wind:** NE at 6 mph, with gusts up to 7 mph\n",
- "\n",
- "### 📅 Today's Forecast\n",
- "- [*](https://www.accuweather.com/en/us/new-york/10021/weather-forecast/14-349727_1_al \"New York City, NY Weather Forecast | AccuWeather\") **High:** 68°F | **Tonight's Low:** 48°F (Clear skies)\n",
- "- [*](https://weather.yahoo.com/us/ny/new-york/ \"New York, NY Weather Forecast, Conditions, and Maps – Yahoo Weather\") Mostly sunny with a high of 69°F and a low of 50°F.\n",
+ "def rng():\n",
+ " return random.random()\n",
"\n",
- "### 🌡️ Hourly Outlook\n",
- "[*](https://www.accuweather.com/en/us/new-york/10021/weather-forecast/14-349727_1_al \"New York City, NY Weather Forecast | AccuWeather\") \n",
- "- 4 PM: 67°F\n",
- "- 5 PM: 64°F\n",
- "- 6 PM: 62°F\n",
- "- 7 PM: 60°F\n",
- "- 8 PM: 58°F\n",
- "- 9 PM: 55°F\n",
- "- 10 PM: 54°F\n",
+ "print(rng())\n",
+ "`````\n",
"\n",
+ "🧠🧠🧠\n",
"\n",
- "### ⚠️ Air Quality & Pollen\n",
- "- [*](https://www.accuweather.com/en/us/new-york/10021/weather-forecast/14-349727_1_al \"New York City, NY Weather Forecast | AccuWeather\") **Air Quality:** Poor\n",
- "- [*](https://www.accuweather.com/en/us/new-york/10021/weather-forecast/349727 \"New York, NY Weather Forecast | AccuWeather\") The air has reached a high level of pollution and is unhealthy for sensitive groups. It is advised to reduce time spent outside if you are feeling symptoms such as difficulty breathing or throat irritation.\n",
- "- [*](https://www.accuweather.com/en/us/new-york/10021/weather-forecast/14-349727_1_al \"New York City, NY Weather Forecast | AccuWeather\") **Tree Pollen:** Moderate | **Grass Pollen:** High | **Mold:** Moderate | **Dust & Dander:** High\n",
"\n",
- "### 🌦️ Looking Ahead\n",
- "[*](https://www.accuweather.com/en/us/new-york/10021/weather-forecast/14-349727_1_al \"New York City, NY Weather Forecast | AccuWeather\") Warmer and dry conditions are expected early this week before cooler air returns for the end of April and early May.\n",
- "- ⏳
web_search(query='weather in NYC today') ⏳)"
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "42)"
]
},
"metadata": {},
@@ -10205,266 +12146,423 @@
}
],
"source": [
- "chat_stream_tools = AsyncChat(model, search='l')\n",
- "res = await chat_stream_tools(\"Search the weather in NYC\", stream=True)\n",
- "_=await adisplay_stream(res)"
+ "for m in ms:\n",
+ " display(Markdown(f'**{m}:**'))\n",
+ " chat = AsyncChat(m, sp=cft_sp, tools=[python])\n",
+ " res = await chat(cft_pr, stream=True)\n",
+ " fmt = await adisplay_stream(res)\n",
+ " try: test(fmt.outp, '`````py', operator.contains)\n",
+ " except AssertionError: display(Markdown(f'**Model {m} not following instructions**')); continue\n",
+ " test(fmt.outp, '`````result\\n42\\n`````', operator.contains) \n",
+ " res = await chat(\"Do it again\", stream=True)\n",
+ " fmt = await adisplay_stream(res)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "ca80fd86",
+ "metadata": {},
+ "source": [
+ "Mix code fences and tool calls"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "070c4dbb",
+ "id": "25924af8",
"metadata": {},
"outputs": [
{
"data": {
"text/markdown": [
- "Hello! How can I help you today?\n",
- "\n",
- "
\n",
+ "\n",
"\n",
- "- id: `chatcmpl-xxx`\n",
- "- model: `claude-haiku-4-5-20251001`\n",
- "- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=12, prompt_tokens=9, total_tokens=21, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=12, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=9, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None)`\n",
+ "**gemini/gemini-3-pro-preview:**\n",
"\n",
- ""
+ "
"
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-haiku-4-5-20251001', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='Hello! How can I help you today?', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=12, prompt_tokens=9, total_tokens=21, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=12, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=9, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None))"
+ "Markdown(**gemini/gemini-3-pro-preview:**)"
]
},
- "execution_count": null,
"metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "chat = AsyncChat(haik45)\n",
- "await chat(\"Hi.\")"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "6bea9f97",
- "metadata": {},
- "outputs": [
+ "output_type": "display_data"
+ },
{
"data": {
"text/markdown": [
"\n",
"\n",
- "Hi! How's it going? How can I help you today?\n",
+ " \n",
"\n",
"
"
],
"text/plain": [
- "Markdown(Hi! How's it going? How can I help you today?)"
+ "Markdown( )"
]
},
"metadata": {},
"output_type": "display_data"
- }
- ],
- "source": [
- "chat = AsyncChat(haik45)\n",
- "res = await chat(\"Hi.\", stream=True)\n",
- "await adisplay_stream(res);"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "0aa1f068",
- "metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠Hi there! How can I help you today?"
- ]
- }
- ],
- "source": [
- "chat = AsyncChat(qwen3p6p)\n",
- "res = await chat(\"Hi.\", stream=True)\n",
- "async for o in res: print(delta_text(o) or '', end='')"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "2538693a",
- "metadata": {},
- "outputs": [
+ },
{
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠"
- ]
+ "data": {
+ "text/markdown": [
+ "\n",
+ "\n",
+ "**Model gemini/gemini-3-pro-preview not following instructions**\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(**Model gemini/gemini-3-pro-preview not following instructions**)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
},
{
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠\n",
- "🔧 simple_add\n",
- "🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠5478954793 + 547982745 = 6026937538"
- ]
- }
- ],
- "source": [
- "chat = AsyncChat(qwen3p6p, tools=[simple_add])\n",
- "res = await chat(\"What's 5478954793+547982745? Use the tool.\", stream=True)\n",
- "async for o in res:\n",
- " if isinstance(o,ModelResponseStream): print(delta_text(o) or '', end='')"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "57adae96",
- "metadata": {},
- "outputs": [],
- "source": [
- "#| export\n",
- "dsf = \"deepseek/deepseek-v4-flash\"\n",
- "dsp = \"deepseek/deepseek-v4-pro\"\n",
- "\n",
- "v4_flash_info = dict(get_model_info(\"deepseek/deepseek-v3.2\"))\n",
- "v4_flash_info |= dict(supports_assistant_prefill=True, supports_function_calling=True, supports_prompt_caching=True,\n",
- " supports_reasoning=True, supports_tool_choice=True)\n",
- "v4_flash_info.update(input_cost_per_token=1.4e-07, input_cost_per_token_cache_hit=2.8e-09, output_cost_per_token=2.8e-07,\n",
- " max_input_tokens=1048576, max_output_tokens=393216, max_tokens=393216)\n",
- "# TODO temp pricing until end of May 2026\n",
- "v4_pro_info = {**v4_flash_info,\n",
- " 'input_cost_per_token': 4.35e-07, 'input_cost_per_token_cache_hit': 3.625e-09, 'output_cost_per_token': 8.7e-07}\n",
- "register_model({dsf: v4_flash_info, dsp: v4_pro_info})\n",
- "get_model_info(dsf);"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "751e1339",
- "metadata": {},
- "outputs": [
+ "data": {
+ "text/markdown": [
+ "\n",
+ "\n",
+ "**gemini/gemini-3-flash-preview:**\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(**gemini/gemini-3-flash-preview:**)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
{
"data": {
"text/markdown": [
- "Hello! How can I assist you today?\n",
+ "\n",
"\n",
- "\n",
"\n",
- "- id: `chatcmpl-xxx`\n",
- "- model: `deepseek-v4-flash`\n",
- "- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=94, prompt_tokens=10, total_tokens=104, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=84, rejected_prediction_tokens=None, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None), prompt_cache_hit_tokens=0, prompt_cache_miss_tokens=10)`\n",
+ "- ⏳ python(code='import random\\nprint(random.randint(1, 10…') ⏳\n",
"\n",
- " "
+ "\n",
+ "python(code='import random\\nprint(random.randint(1, 10…')→42 \n",
+ "\n",
+ "\n",
+ "\n",
+ "`````py\n",
+ "import random\n",
+ "\n",
+ "def get_another_random():\n",
+ " return random.randint(1, 100)\n",
+ "\n",
+ "print(get_another_random())\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "The first random number generated was 42 and the second random number generated was 42.\n",
+ "\n",
+ "
"
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='deepseek-v4-flash', object='chat.completion', system_fingerprint='fp_058df29938_prod0820_fp8_kvcache_20260402', choices=[Choices(finish_reason='stop', index=0, message=Message(content='Hello! How can I assist you today?', role='assistant', tool_calls=None, function_call=None, reasoning_content='Hmm, the user just greeted me with a simple \"Hi!\" This is a very common opening. They might be testing if I\\'m responsive, or they could be about to ask a question. Since there\\'s no specific request yet, the best approach is to return the greeting warmly and invite them to state their needs. A friendly, open-ended response keeps the conversation flowing. I\\'ll keep it simple and positive.', provider_specific_fields={'reasoning_content': 'Hmm, the user just greeted me with a simple \"Hi!\" This is a very common opening. They might be testing if I\\'m responsive, or they could be about to ask a question. Since there\\'s no specific request yet, the best approach is to return the greeting warmly and invite them to state their needs. A friendly, open-ended response keeps the conversation flowing. I\\'ll keep it simple and positive.'}), provider_specific_fields={})], usage=Usage(completion_tokens=94, prompt_tokens=10, total_tokens=104, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=84, rejected_prediction_tokens=None, text_tokens=None, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=None, image_tokens=None, video_tokens=None), prompt_cache_hit_tokens=0, prompt_cache_miss_tokens=10))"
+ "Markdown(\n",
+ "- ⏳ python(code='import random\\nprint(random.randint(1, 10…') ⏳\n",
+ "\n",
+ "\n",
+ "python(code='import random\\nprint(random.randint(1, 10…')→42 \n",
+ "\n",
+ "\n",
+ "\n",
+ "`````py\n",
+ "import random\n",
+ "\n",
+ "def get_another_random():\n",
+ " return random.randint(1, 100)\n",
+ "\n",
+ "print(get_another_random())\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "The first random number generated was 42 and the second random number generated was 42.)"
]
},
- "execution_count": null,
"metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "completion(dsf, [mk_msg(\"Respond in English.\", role='system'), mk_msg(\"Hi!\")])"
- ]
- },
- {
- "cell_type": "markdown",
- "id": "5f4d05f9",
- "metadata": {},
- "source": [
- "### Tool Call Referencing"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "2e88c6e6",
- "metadata": {},
- "outputs": [
+ "output_type": "display_data"
+ },
{
"data": {
"text/markdown": [
- "Perfect! I successfully:\n",
- "1. Called `get_person` which returned Alice's information (name: Alice, age: 30)\n",
- "2. Passed that result to `greet_person` which greeted her with: \"Hello Alice, you are 30 years old!\"\n",
+ "\n",
"\n",
- "\n",
+ "**claude-sonnet-4-6:**\n",
"\n",
- "- id: `chatcmpl-xxx`\n",
- "- model: `claude-sonnet-4-5-20250929`\n",
- "- finish_reason: `stop`\n",
- "- usage: `Usage(completion_tokens=64, prompt_tokens=1024, total_tokens=1088, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=64, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=1024, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None)`\n",
+ "
"
+ ],
+ "text/plain": [
+ "Markdown(**claude-sonnet-4-6:**)"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/markdown": [
+ "\n",
"\n",
- ""
+ "Sure! Let's start by calling the `python` tool directly to generate the first random number.\n",
+ "- ⏳ python(code='import random\\nprint(random.randint(1, 10…') ⏳\n",
+ "\n",
+ "\n",
+ "python(code='import random\\nprint(random.randint(1, 10…')→42 \n",
+ "\n",
+ "The first number is **42**. Now let's run the fenced code block to generate the second number:\n",
+ "\n",
+ "`````py\n",
+ "import random\n",
+ "\n",
+ "def random_in_range(low, high):\n",
+ " \"\"\"Return a random integer between low and high, inclusive.\"\"\"\n",
+ " return random.randint(low, high)\n",
+ "\n",
+ "print(random_in_range(1, 100))\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠🧠🧠🧠🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "Here's a quick summary of both results:\n",
+ "\n",
+ "- **First number** (generated via direct `python` tool call): **42**\n",
+ "- **Second number** (generated via the 5-backtick `py` fenced code block): **42**\n",
+ "\n",
+ "Quite the coincidence — both methods landed on **42**! 🎲 Of course, this was just random chance; running either method again would almost certainly produce different results across the 1–100 range.\n",
+ "\n",
+ "
"
],
"text/plain": [
- "ModelResponse(id='chatcmpl-xxx', created=1000000000, model='claude-sonnet-4-5-20250929', object='chat.completion', system_fingerprint=None, choices=[Choices(finish_reason='stop', index=0, message=Message(content='Perfect! I successfully:\\n1. Called `get_person` which returned Alice\\'s information (name: Alice, age: 30)\\n2. Passed that result to `greet_person` which greeted her with: \"Hello Alice, you are 30 years old!\"', role='assistant', tool_calls=None, function_call=None, provider_specific_fields={'citations': None, 'thinking_blocks': None}))], usage=Usage(completion_tokens=64, prompt_tokens=1024, total_tokens=1088, completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None, audio_tokens=None, reasoning_tokens=0, rejected_prediction_tokens=None, text_tokens=64, image_tokens=None, video_tokens=None), prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=0, text_tokens=1024, image_tokens=None, video_tokens=None, cache_creation_tokens=0, cache_creation_token_details=CacheCreationTokenDetails(ephemeral_5m_input_tokens=0, ephemeral_1h_input_tokens=0)), cache_creation_input_tokens=0, cache_read_input_tokens=0, inference_geo='not_available', speed=None))"
+ "Markdown(Sure! Let's start by calling the `python` tool directly to generate the first random number.\n",
+ "- ⏳ python(code='import random\\nprint(random.randint(1, 10…') ⏳\n",
+ "\n",
+ "\n",
+ "python(code='import random\\nprint(random.randint(1, 10…')→42 \n",
+ "\n",
+ "The first number is **42**. Now let's run the fenced code block to generate the second number:\n",
+ "\n",
+ "`````py\n",
+ "import random\n",
+ "\n",
+ "def random_in_range(low, high):\n",
+ " \"\"\"Return a random integer between low and high, inclusive.\"\"\"\n",
+ " return random.randint(low, high)\n",
+ "\n",
+ "print(random_in_range(1, 100))\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠🧠🧠🧠🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "Here's a quick summary of both results:\n",
+ "\n",
+ "- **First number** (generated via direct `python` tool call): **42**\n",
+ "- **Second number** (generated via the 5-backtick `py` fenced code block): **42**\n",
+ "\n",
+ "Quite the coincidence — both methods landed on **42**! 🎲 Of course, this was just random chance; running either method again would almost certainly produce different results across the 1–100 range.)"
]
},
- "execution_count": null,
"metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "achat = AsyncChat('claude-sonnet-4-5', tools=[get_person, greet_person], tc_refs=True)\n",
- "await achat(\"First call get_person, then pass the result to greet_person\", max_steps=3)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "ba39edd3",
- "metadata": {},
- "outputs": [
+ "output_type": "display_data"
+ },
{
"data": {
+ "text/markdown": [
+ "\n",
+ "\n",
+ "**openai/gpt-5.4:**\n",
+ "\n",
+ "
"
+ ],
"text/plain": [
- "{'toolu_01VrvqXwSeiAqi9KXEkt2svo': {'name': 'Alice', 'age': 30},\n",
- " 'toolu_019msmJ6FxNCUeH3DsJdhd4e': 'Hello Alice, you are 30 years old!'}"
+ "Markdown(**openai/gpt-5.4:**)"
]
},
- "execution_count": null,
"metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "achat.tc_res"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "ecfd8af3",
- "metadata": {},
- "outputs": [
+ "output_type": "display_data"
+ },
{
"data": {
+ "text/markdown": [
+ "\n",
+ "\n",
+ "\n",
+ "- ⏳ python(code='import random\\nprint(random.randint(1,100…') ⏳\n",
+ "\n",
+ "\n",
+ "python(code='import random\\nprint(random.randint(1,100…')→42 \n",
+ "\n",
+ "`````py\n",
+ "import random\n",
+ "\n",
+ "def second_random():\n",
+ " return random.randint(1, 100)\n",
+ "\n",
+ "print(second_random())\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "Both numbers were 42 and 42.\n",
+ "\n",
+ "
"
+ ],
"text/plain": [
- "[[ChatCompletionMessageToolCall(index=1, caller={'type': 'direct'}, function=Function(arguments='{}', name='get_person'), id='toolu_01VrvqXwSeiAqi9KXEkt2svo', type='function')],\n",
- " [ChatCompletionMessageToolCall(index=1, caller={'type': 'direct'}, function=Function(arguments='{\"person\": \"$`toolu_01VrvqXwSeiAqi9KXEkt2svo`\"}', name='greet_person'), id='toolu_019msmJ6FxNCUeH3DsJdhd4e', type='function')]]"
+ "Markdown(\n",
+ "- ⏳ python(code='import random\\nprint(random.randint(1,100…') ⏳\n",
+ "\n",
+ "\n",
+ "python(code='import random\\nprint(random.randint(1,100…')→42 \n",
+ "\n",
+ "`````py\n",
+ "import random\n",
+ "\n",
+ "def second_random():\n",
+ " return random.randint(1, 100)\n",
+ "\n",
+ "print(second_random())\n",
+ "`````\n",
+ "\n",
+ "🧠🧠🧠\n",
+ "\n",
+ "\n",
+ "`````result\n",
+ "42\n",
+ "`````\n",
+ "Both numbers were 42 and 42.)"
]
},
- "execution_count": null,
"metadata": {},
- "output_type": "execute_result"
+ "output_type": "display_data"
}
],
"source": [
- "list(L(achat.hist).attrgot('tool_calls').filter())"
+ "pr = 'First, directly call the `python` tool from your available tool schema to print a random integer between 1 and 100. After you receive that tool result, write a 5-backtick `py` code fence that defines a second random-number function and prints one call to it, so it can be executed as a fenced code tool. After receiving the fenced result, briefly report both numbers.'\n",
+ "for m in ms:\n",
+ " display(Markdown(f'**{m}:**'))\n",
+ " chat = AsyncChat(m, sp=cft_sp, tools=[python])\n",
+ " res = await chat(pr, stream=True)\n",
+ " fmt = await adisplay_stream(res)\n",
+ " try: test(fmt.outp, tool_dtls_tag, operator.contains)\n",
+ " except AssertionError: display(Markdown(f'**Model {m} not following instructions**')); continue\n",
+ " test(fmt.outp, '`````result\\n42\\n`````', operator.contains)"
]
},
{
@@ -10584,12 +12682,12 @@
"text/markdown": [
"\n",
"\n",
- "Hi Jeremy — nice to meet you. How can I help?\n",
+ "Hi Jeremy! How can I help you today?\n",
"\n",
"
"
],
"text/plain": [
- "Markdown(Hi Jeremy — nice to meet you. How can I help?)"
+ "Markdown(Hi Jeremy! How can I help you today?)"
]
},
"metadata": {},
@@ -10810,7 +12908,7 @@
"\n",
"```json\n",
"{\n",
- " \"id\": \"call_vJtw0mTMy2AhjbA36LcyaXNF\",\n",
+ " \"id\": \"call_JonQUnAOqtGJcAn8jzTLYNPr\",\n",
" \"call\": {\n",
" \"function\": \"simple_add\",\n",
" \"arguments\": {\n",
@@ -10831,7 +12929,7 @@
"\n",
"```json\n",
"{\n",
- " \"id\": \"call_DUwGqBY80min0Po9fMLDAWmM\",\n",
+ " \"id\": \"call_DTAxK90n2ypFzQ0y3nssCbfz\",\n",
" \"call\": {\n",
" \"function\": \"simple_add\",\n",
" \"arguments\": {\n",
@@ -10853,7 +12951,7 @@
"\n",
"```json\n",
"{\n",
- " \"id\": \"call_p80UAoYsrQAolFJ8DT5UPUKv\",\n",
+ " \"id\": \"call_3hGwl9tKxwDSYMUKOsb8dDZh\",\n",
" \"call\": {\n",
" \"function\": \"multiply\",\n",
" \"arguments\": {\n",
@@ -10867,7 +12965,7 @@
"\n",
" \n",
"\n",
- "\\((5+7)\\times(5+6)=12\\times11=132\\)\n",
+ "\\((5+7)(5+6) = 12 \\times 11 = 132\\)\n",
"\n",
"
\n",
"\n",
- "This is a **domestic dog** — likely a **Cavalier King Charles Spaniel puppy**.\n",
+ "Dog (a puppy, likely a Cavalier King Charles Spaniel).\n",
"\n",
"
"
],
"text/plain": [
- "Markdown(This is a **domestic dog** — likely a **Cavalier King Charles Spaniel puppy**.)"
+ "Markdown(Dog (a puppy, likely a Cavalier King Charles Spaniel).)"
]
},
"metadata": {},
@@ -11049,20 +13147,26 @@
"text/markdown": [
"\n",
"\n",
- "🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠\n",
+ "🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠\n",
"\n",
- "Today in Brisbane, Queensland: **Sunny now, about 71°F / 21°C**.\n",
+ "Today in **Brisbane, Queensland — May 5, 2026**:\n",
"\n",
- "Forecast for the rest of today: mostly **cloudy**, warming to around **82°F / 28°C** early afternoon, then cooling to about **68°F / 20°C** tonight.\n",
+ "- **Current:** Cloudy, about **69°F / 20°C**\n",
+ "- **Morning:** Cloudy early, becoming mostly sunny; **63–75°F / 17–24°C**\n",
+ "- **Midday:** Partly sunny to intermittent clouds; around **78–79°F / 25–26°C**\n",
+ "- **Afternoon:** Chance of showers, warming to about **81°F / 27°C**\n",
"\n",
"
"
],
"text/plain": [
- "Markdown(🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠\n",
+ "Markdown(🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠\n",
"\n",
- "Today in Brisbane, Queensland: **Sunny now, about 71°F / 21°C**.\n",
+ "Today in **Brisbane, Queensland — May 5, 2026**:\n",
"\n",
- "Forecast for the rest of today: mostly **cloudy**, warming to around **82°F / 28°C** early afternoon, then cooling to about **68°F / 20°C** tonight.)"
+ "- **Current:** Cloudy, about **69°F / 20°C**\n",
+ "- **Morning:** Cloudy early, becoming mostly sunny; **63–75°F / 17–24°C**\n",
+ "- **Midday:** Partly sunny to intermittent clouds; around **78–79°F / 25–26°C**\n",
+ "- **Afternoon:** Chance of showers, warming to about **81°F / 27°C**)"
]
},
"metadata": {},
@@ -11129,7 +13233,7 @@
{
"data": {
"text/plain": [
- "total=3,346 | in=3,164 | out=182 | cached=85.0% | reasoning=78 | searches=0 | $0.0046"
+ "total=3,322 | in=3,124 | out=198 | cached=86.0% | reasoning=92 | searches=0 | $0.0047"
]
},
"execution_count": null,
@@ -11215,7 +13319,14 @@
"text": [
"\n",
"\n",
- "🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠31304616"
+ "🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠🧠31304616"
]
}
],
@@ -11341,9 +13452,9 @@
"metadata": {
"solveit": {
"default_code": false,
- "mode": "concise",
+ "mode": "learning",
"use_fence": false,
- "use_thinking": false,
+ "use_thinking": true,
"use_tools": true,
"ver": 2
}