BerriAI / litellm

Python SDK, Proxy Server (LLM Gateway) to call 100+ LLM APIs in OpenAI format - [Bedrock, Azure, OpenAI, VertexAI, Cohere, Anthropic, Sagemaker, HuggingFace, Replicate, Groq]
https://docs.litellm.ai/docs/
Other
13.66k stars 1.6k forks source link

[Feature]: Automate updating our list of HF TGI and Conversational models #473

Open ishaan-jaff opened 1 year ago

ishaan-jaff commented 1 year ago

The Feature

How we currently get this list: https://github.com/BerriAI/litellm/blob/main/cookbook/get_hf_models.py

This should be auto-updated: https://github.com/BerriAI/litellm/tree/main/litellm/llms/huggingface_llms_metadata

This is done manually, it should be automated on a weekly basis

Motivation, pitch

-

Twitter / LinkedIn details

No response

linediconsine commented 1 year ago

Hi ishaan-jaff how do you suggest to automate it? Whats the plan?

Thanks

ishaan-jaff commented 1 year ago

@linediconsine not entirely sure as yet. For now I just plan on running this script on a weekly basis to update the model list

ishaan-jaff commented 1 year ago

The better way to do this would be to have some type of job that runs