langchain integration package langchain-localai

[mkhludnev]

While working on one langchain application I need high performance embeddings, which I've found in LocalAI. It can run embeddings on GPU and provides them in bulks!
Unfortunately, LocalAIEmbeddings in langchain is not able to request bulks, and depends on outdated openai sdk v0. I attempt to fix it in langchain-ai/langchain#22399 and langchain-ai/langchain#22666, but was advised to spit off a standalone integration package. Here we go:

• source under my personal acct https://github.com/mkhludnev/langchain-localai
• releases https://pypi.org/project/langchain-localai/#history
• reference from Langchain docs: https://python.langchain.com/docs/integrations/text_embedding/localai/

I'm going to maintain this project under my account. And I'm ok to contribute this repository here to @mudler organization\account or other (even without committer permission - I'm ok with it).
So, please let me know if you wish to migrate https://github.com/mkhludnev/langchain-localai to another org.
Thanks.