Use Bedrock embedding model via litellm?

[gmerritt]

The title sums it up, but I posted this just now on the rag_api repo discussions:

danny-avila/rag_api#158

I'm echoing it here, as the LibreChat discussion seems to have a lot more activity. :)

If anyone has advice, please share either here or (maybe better?) on the rag_api discussion.

Thanks!

[gmerritt]

Thanks! And oh right -- I got my vector embeddings working, I should add my solution to discussions for completeness. Thanks, @onestardao !

[gmerritt]

This environment variable combination works for me!

EMBEDDINGS_PROVIDER=openai
EMBEDDINGS_MODEL=amazon.titan-embed-text-v2:0
RAG_OPENAI_API_KEY=sk-<litellm_key>
RAG_OPENAI_BASEURL=https://my.litellm.hostname/v1

The "trick" was wrapping my head around LiteLLM standardizing to present everything as OpenAI...even if (as in this example) the true endpoint is Bedrock.