Can you also proxy to embeddings endpoint?

[kengoodridge]

Could you find the time to proxy to embeddings, supporting openai api, specifically for ollama, but I would guess you should do it for any of the llm providers that support it. I don't know how general embeddings support is.

I think it would be particularly useful and clean for RAG.

Thanks! Find this package invaluable.

[ishaan-jaff]

Hi @kengoodridge thanks for the issue will work on this today

here's our supported embedding models btw: https://docs.litellm.ai/docs/embedding/supported_embedding

[krrishdholakia]

This is now live - @kengoodridge

litellm/litellm/proxy/proxy_server.py

Line 607 in 8291f23

async def embeddings(request: Request):