Add Google Ai studio API config

[the-muzammilali]

Add option to let users use Free Gemini Ai studio API keys.

[a-holm]

You already can, I tested it. There was a merged pull request today that fixes some bugs with it, but it was updated a few hours ago. Just put something like this in the /config/config.toml file:

#Global LLM configuration
[llm]
model = "gemini-2.0-flash"
base_url = "https://generativelanguage.googleapis.com/v1beta/openai/"
api_key = "YOUR-API-KEY"
max_tokens = 4096
temperature = 0.0

the base url and so on is taken from here https://ai.google.dev/gemini-api/docs/openai#python

for available models check out:

https://ai.google.dev/gemini-api/docs/models/gemini
and
https://ai.google.dev/gemini-api/docs/models/experimental-models

not all of them work, but two of the ones that do work are:

gemini-2.0-flash
gemini-2.0-pro-exp-02-05

at least, the thinking models do not work because they can't handle function calling. I havent tes

Although with the experimental gemini-2.0-pro-exp-02-05 you might need to add a 6-7 second pause between each prompt being made because it has very restrictive rate limits. gemini-2.0-flash works fine, but it is a little bit dumb so it sometimes uses tools wrong (misinterprets what it reads from the result) or misinterpret what you say and so on. It often works though.