Skip to content

docs(models): document litellm-sdk embeddings provider#1336

Merged
nicoloboschi merged 2 commits into
vectorize-io:mainfrom
r266-tech:docs-embeddings-litellm-sdk-provider
Apr 30, 2026
Merged

docs(models): document litellm-sdk embeddings provider#1336
nicoloboschi merged 2 commits into
vectorize-io:mainfrom
r266-tech:docs-embeddings-litellm-sdk-provider

Conversation

@r266-tech

Copy link
Copy Markdown
Contributor

Summary

litellm-sdk is a real, functional embeddings provider (registered via embeddings_litellm_sdk_* config in hindsight-api-slim/hindsight_api/config.py, instantiated by the LiteLLMSDKEmbeddings class at embeddings.py:753, and listed in the unknown-provider error message: "Supported: 'local', 'tei', 'openai', 'cohere', 'google', 'litellm', 'litellm-sdk'" at embeddings.py:1209). It also gets continued investment — #1320 (merged a few hours ago) wired allowed_openai_params=["dimensions"] so HINDSIGHT_API_EMBEDDINGS_LITELLM_SDK_OUTPUT_DIMENSIONS actually works for OpenAI-compatible custom models.

But the embedding providers table at docs/developer/models.mdx only lists six providers (local / openai / cohere / google / tei / litellm) — litellm-sdk is missing. The cross-encoder providers table on the same page already documents it (| litellm-sdk | LiteLLM SDK (direct API, no proxy) | Multi-provider, simpler setup |), so users reading the embeddings section can't tell the SDK path is supported.

What's in the PR

  • hindsight-docs/docs/developer/models.mdx (+5 LOC): one new row in the embedding providers table mirroring the cross-encoder description; one new # LiteLLM SDK (direct, no proxy) stanza in the Configuration Examples block (mirrors the existing # LiteLLM proxy stanza style).
  • skills/hindsight-docs/references/developer/models.md (+5 LOC): byte-for-byte mirror.

Source of truth

  • Provider name: LiteLLMSDKEmbeddings.provider_name returns "litellm-sdk" (embeddings.py:802); factory dispatches provider == "litellm-sdk" (embeddings.py:1173).
  • Env var names: HINDSIGHT_API_EMBEDDINGS_LITELLM_SDK_API_KEY and HINDSIGHT_API_EMBEDDINGS_LITELLM_SDK_MODEL defined at config.py:223-224. API key is required (factory raises ValueError if unset, embeddings.py:1175-1178).
  • Example model openai/text-embedding-3-small follows the LiteLLM provider-prefix convention used throughout the existing docs (and is the path fix(embeddings): add allowed_openai_params for OpenAI-compatible embedding dimensions #1320 just made fully functional).

Pure docs, no behavior change.

@nicoloboschi nicoloboschi left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@nicoloboschi nicoloboschi merged commit 4282e84 into vectorize-io:main Apr 30, 2026
51 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants