fix: use custom_llm_provider from kwargs if provided

[lwcobo]

Title

In litellm.main.aembedding, use custom_llm_provider from kwargs if provided

Relevant issues

For locally deployed embedding models, custom_llm_provider is explicitly provided via kwargs, and it's not possible to infer the provider from the model name.

Pre-Submission checklist

N/A

Type

🐛 Bug Fix

Changes

### EMBEDDING ENDPOINTS ####################
@client
async def aembedding(*args, **kwargs) -> EmbeddingResponse:
    ...
    _, custom_llm_provider, _, _ = get_llm_provider(
        model=model, 
        custom_llm_provider=kwargs.get("custom_llm_provider", None),
        api_base=kwargs.get("api_base", None),
    )
    ...

[vercel]

The latest updates on your projects. Learn more about Vercel for Git ↗︎

Name	Status	Preview	Comments	Updated (UTC)
litellm	✅ Ready (Inspect)	Visit Preview	💬 Add feedback	Apr 2, 2025 6:24am