-
Notifications
You must be signed in to change notification settings - Fork 1.9k
Description
What specific problem does this solve?
Currently, to use Hugging Face models, you need to use the OpenAI compatible provider option, which can bring complexity, specially regarding different the providers and models Hugging Face itself provides.
Additional context (optional)
No response
Roo Code Task Links (Optional)
No response
Request checklist
- I've searched existing Issues and Discussions for duplicates
- This describes a specific problem with clear impact and context
Interested in implementing this?
- Yes, I'd like to help implement this feature
Implementation requirements
- I understand this needs approval before implementation begins
How should this be solved? (REQUIRED if contributing, optional otherwise)
I already created an implementation, apologies. It is available over here https://github.com/TGlide/Roo-Code/tree/hf-provider
How will we know it works? (Acceptance Criteria - REQUIRED if contributing, optional otherwise)
Given I have configured the Hugging Face provider and model in Roo's settings
When I try and chat with said model
Then I should see the model responds correctly
But it does not error ou
Technical considerations (REQUIRED if contributing, optional otherwise)
For implementation, I went off the OpenAI compatible provider implementation, but tailored it specifically to Hugging Face's router URL. In the .tsx file for it, I also fetch the models, and cache them locally to avoid having to refetch said models.
Trade-offs and risks (REQUIRED if contributing, optional otherwise)
Alternative approach was to not fetch said models and just allow the user to but the model and provider manually, but that is more cumbersome and beats the purpose of having said provider IMO
Metadata
Metadata
Assignees
Labels
Type
Projects
Status