fix: replace delisted x-ai/grok-4-fast default with the live openai/gpt-4.1-mini#246
Closed
Fearvox wants to merge 668 commits into
Closed
fix: replace delisted x-ai/grok-4-fast default with the live openai/gpt-4.1-mini#246Fearvox wants to merge 668 commits into
Fearvox wants to merge 668 commits into
Conversation
Use self-deployed embedding and rerank APIs by default See merge request npc-work/aic/ai/evermemos-opensource!64
vLLM Rerank API adopts an instruction-tuned approach See merge request npc-work/aic/ai/evermemos-opensource!65
feat: metrics client and rerank/vectorize/retrieve metrics See merge request npc-work/aic/ai/evermemos-opensource!66
fix:update episode prompt See merge request npc-work/aic/ai/evermemos-opensource!68
feat: add rerank metrics See merge request npc-work/aic/ai/evermemos-opensource!69
* chore: rename project from evermemos to EverCore This commit renames the project directory and updates all internal references from "evermemos" to "EverCore". The changes include: - Renaming the main directory from `methods/evermemos` to `methods/EverCore` - Updating all import paths and module references - Maintaining the same code structure and functionality - Adding new configuration files (.vscode/settings.json, .pylintrc, pyrightconfig.json) - Updating Dockerfile and project metadata * docs: update references from evermemos to EverCore Update documentation files to reflect the renaming of the 'evermemos' directory to 'EverCore'. This includes fixing clone commands, directory paths, and documentation links across multiple files to ensure consistency and correct navigation for users. * chore: rename EverMemOS to EverCore across codebase This is a project-wide rebranding from EverMemOS to EverCore. The changes include: - Update project name in source files, documentation, and configuration - Rename API references, environment variables, and service names - Modify demo descriptions and benchmark configurations - Update URLs and citations to reflect new project identity All functionality remains identical; only naming has changed to align with the new project branding. * docs: update README with EverCore focus and restructured TOC - Add line break before Table of Contents for better visual separation - Rewrite project description to highlight EverCore as the central component - Reorder directory tree to prioritize benchmarks and methods over use-cases - Update use-cases list with more examples and clarify they are templates - Improve flow from Quick Start to use-cases to benchmarks * docs: update README with clearer methods description and benchmarks Add benchmark numbers directly in the method summaries for better visibility. Clarify introductory text to emphasize choice and composition of methods. * docs: fix markdown formatting in README table of contents Adjust whitespace and line breaks to ensure proper rendering of the collapsible table of contents section.
…d-AI#204) - Replace specific EverMemBench-Dynamic badge with general EverMind-AI HuggingFace badge - Remove redundant License badge - Change "Methods" section heading to "Architecture Methods" - Update sub-section headings from h4 (####) to h3 (###) for better hierarchy
…rMind-AI#208) * docs: restructure README and add AGENTS.md for better navigation - Reorder sections to emphasize architecture methods and use cases - Move use cases section before quick start for better flow - Rename "Methods" to "Architecture Methods" for clarity - Add AGENTS.md with quick commands and key entry points - Update section headers to improve document hierarchy - Maintain all existing content while improving organization * docs: add community and contribution files * docs: reorder README directory tree for logical grouping * docs: move community files to .github/ and update references * ci: change deploy workflow trigger from feature branch to main
* docs: restructure README and add subdirectory guides Move the directory tree from the main README to new dedicated README files for each top-level folder (use-cases, methods, benchmarks). Add detailed introductions and tables to guide users to the appropriate subprojects. This improves navigation and provides clear entry points for different use cases. * docs: expand showcase section with new projects and links Add six new project entries to the README showcase, each with a banner image, description, and code/plugin link. Also update an existing benchmark entry to include a dataset link. This enhances the repository's demonstration of real-world applications and available resources.
* docs(readme): update project links and formatting * docs(use-cases): enhance README with visual catalogue of demos Expand the use cases section from a simple table to a detailed visual catalogue with project banners, descriptions, and links. This improves user engagement and provides a better showcase of community integrations and demos. * docs: update READMEs and add validation for use-case links
* docs: update plugin repository link in README * docs(readme): update banner gif link
* docs(readme): add four new use case entries * docs(readme): update outdated banner links to correct github repos
…e-demo-content-payload Fix EverCore demo memory payload
…actions-hygiene Harden GitHub Actions workflows
…bot-config Add Dependabot configuration
…adme-quickstart docs: verify EverCore quickstart path
…I#236) Delete deprecated EvoAgentBench, EverMemBench benchmark suites and HyperMem memory system implementation, including all associated configurations, scripts, and supporting assets.
…pt-4.1-mini The shipped default LLM model id is delisted on OpenRouter, so a fresh setup (copy env.template, add a real OPENROUTER_API_KEY, run) fails on the first LLM call. x-ai/grok-4-fast is absent from the OpenRouter catalog and its endpoints list is empty (0 serving providers); llm_provider.py forwards LLM_MODEL verbatim, so the dead default reaches OpenRouter unchanged. Replace it with openai/gpt-4.1-mini, which is the code's own DEFAULT_LLM_MODEL (llm_provider.py:9) and is live on OpenRouter, in the three places the dead id is shipped: env.template (runtime default), docs/dev_docs/getting_started.md (setup example), and docs/usage/CONFIGURATION_GUIDE.md (cost-effective-model example). Template default now agrees with the code default. Leaves tests/test_llm_switching_e2e.py untouched: it uses x-ai/grok-4-fast on purpose as a disallowed model to assert white-list rejection, and a delisted id is still a valid "disallowed" example there. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
The out-of-box LLM model default is a delisted OpenRouter model, so a new
user who copies
env.template, fills in a realOPENROUTER_API_KEY, and runsEverCore fails on the first LLM call.
x-ai/grok-4-fastis no longer served on OpenRouter:GET https://openrouter.ai/api/v1/models— 343 models; the onlygrokids arex-ai/grok-build-0.1,x-ai/grok-4.3,x-ai/grok-4.20,x-ai/grok-4.20-multi-agent).GET …/models/x-ai/grok-4-fast/endpointsreturns 0 serving endpoints, soa chat-completions request with it gets no providers and fails.
src/memory_layer/llm/llm_provider.py:45forwardsLLM_MODELverbatim, so thetemplate value reaches OpenRouter unchanged.
This PR replaces the dead id with
openai/gpt-4.1-mini— which is the code'sown hardcoded
DEFAULT_LLM_MODEL(llm_provider.py:9) and has live OpenRouterendpoints — in the three places the dead id is shipped:
methods/EverCore/env.template(the runtime default)methods/EverCore/docs/dev_docs/getting_started.md(setup example)methods/EverCore/docs/usage/CONFIGURATION_GUIDE.md(the cost-effective-model example)After this change the template default agrees with the code default, so a
fresh
key-only setup works on the first call.Why this id
openai/gpt-4.1-miniis alreadyDEFAULT_LLM_MODELin the code, is live onOpenRouter, and is cost-effective — so template, docs, and code all agree.
Scope
Three files, one dead-id → live-id swap each. No code change. Intentionally
not touched:
tests/test_llm_switching_e2e.pyusesx-ai/grok-4-fastdeliberately as a disallowed model to assert white-list rejection — a delisted
id is still a valid "disallowed" example there, so its meaning is preserved.
Verification
OpenRouter catalog checked live (2026-06-02):
x-ai/grok-4-fastnot present;openai/gpt-4.1-minipresent. A running EverCore configured withopenai/gpt-4.1-miniserves normally.Co-Authored-By: Claude Opus 4.8 (1M context) noreply@anthropic.com
🤖 Generated with Claude Code