Enhancement: Add Ability to Modify AI Providers/Models in Running LibreChat Instance - Githubissues

danny-avila / LibreChat

Enhanced ChatGPT Clone: Features Anthropic, AWS, OpenAI, Assistants API, Azure, Groq, o1, GPT-4o, Mistral, OpenRouter, Vertex AI, Gemini, Artifacts, AI model switching, message search, langchain, DALL-E-3, ChatGPT Plugins, OpenAI Functions, Secure Multi-User System, Presets, completely open-source for self-hosting. Actively in public development.

https://librechat.ai/

MIT License

19.41k stars 3.24k forks source link

Enhancement: Add Ability to Modify AI Providers/Models in Running LibreChat Instance #4674

Closed TedLin1993 closed 2 weeks ago

TedLin1993 commented 2 weeks ago

What features would you like to see added?

Currently, to add or modify models in LibreChat, one must:

Modify Environment Variables: Some models require changes in the .env file.
Edit Configuration Files: Certain configurations need to be set in librechat.yaml.
Redeploy the Application: After making these changes, the entire application needs to be redeployed to use the new models.

This process is quite cumbersome and not user-friendly:

User Experience: It disrupts workflow as users need to stop the service, change configurations, and then restart or redeploy the application.
Flexibility: Users cannot dynamically adjust or test new models without going through this cycle.

Proposed Feature:

I propose adding functionality where:

Frontend Configuration: Users can add or modify AI models directly from the LibreChat frontend interface without needing to dive into configuration files or environment variables.
Persistent Changes: These modifications should persist through redeployments, ensuring that the settings are not lost when the application is updated or restarted.

Benefits:

Improved UX: Allows for real-time adjustments to AI providers and models, enhancing experimentation and usability.
Simplification: Reduces the technical barrier for users who might not be familiar with server-side configurations or deployments.

More details

Possible Implementation:

New Config Table in DB: Introduce a new config table in the database. When this table is empty:
Use the existing settings from .env and librechat.yaml files.
Automatically populate this table with the current configurations from these files.
Prioritize Config Table: If the config table is not empty, the application should use the settings stored in this table for AI providers and models, overriding any settings in .env or librechat.yaml.

Which components are impacted by your request?

No response

Pictures

No response

Code of Conduct

[X] I agree to follow this project's Code of Conduct

danny-avila commented 2 weeks ago

Something like this is planned, closing in favor of https://github.com/danny-avila/LibreChat/issues/3137