[Feature]: add litellm embedding to langchain

[thiswillbeyourgithub]

The Feature

I noticed langchain has a litellm backend for chat but not for embeddings.

Motivation, pitch

I'm coding DocToolsLLM and would like to add support for mistralai instead of OpenAI. But I would preferably use litellm for maximum customizability for the users (changing the LLM model to use with a simple argument change)

Also, mistralai does not yet have embeddings support in langchain and I would prefer to use mistral via litellm than directly

Twitter / LinkedIn details

No response

[thiswillbeyourgithub]

Up: I would still love that

[ishaan-jaff]

hi @thiswillbeyourgithub - thanks for being an active user / contributor to LiteLLM. I'd love to get on a call and learn how we can improve LiteLLM for you + prioritize your issues. What's the best email to send an invite to ?

Here's my cal if that's easier: https://calendly.com/d/4mp-gd3-k5k/litellm-1-1-onboarding-chat

My linkedin if you want to DM : https://www.linkedin.com/in/reffajnaahsi/

[thiswillbeyourgithub]

Here's my cal if that's easier: https://calendly.com/d/4mp-gd3-k5k/litellm-1-1-onboarding-chat

Done

[pazevedo-hyland]

Any updates on this? @thiswillbeyourgithub / @ishaan-jaff

[github-actions[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs.

[thiswillbeyourgithub]

I ended up doing my own litellm embeddings class and it seems to be working fine: https://github.com/thiswillbeyourgithub/wdoc/blob/main/wdoc/utils/customs/litellm_embeddings.py

[github-actions[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs.