feat: support multiple text embedding providers #76
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
donghaoren
reviewed
Oct 28, 2025
packages/backend/pyproject.toml
Outdated
| "llvmlite >= 0.43.0", | ||
| "accelerate >= 1.5.0", | ||
| "tqdm >= 4.60.0", | ||
| "litellm>=1.78.5", |
There was a problem hiding this comment.
Suggested change
| "litellm>=1.78.5", | |
| "litellm >= 1.78.5", |
| text_projector = _project_text_with_sentence_transformers | ||
|
|
||
| hasher = Hasher() | ||
| hasher.update( |
There was a problem hiding this comment.
In _projection_for_texts, we have a caching mechanism that computes a hash from the parameters and use the hash as a cache filename. It seems like we are not taking into account all text projector parameters (e.g., the text projector type, dimensions). Could you update the code to include the text projector type (as string) and args into the hasher, and increment the "version" number?
There was a problem hiding this comment.
Of course. Thanks for pointing it out. Addressed this in 823a82b.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary
Adds support for multiple text embedding providers through LiteLLM integration, enabling users to leverage API-based models (OpenAI, Cohere, Azure, Ollama, etc.) alongside the existing local SentenceTransformers approach.
compute_text_projection()API remains 100% backward compatible; new parameters are optional.Changes
--text-projectorto route betweenlitellmandsentence_transformers, and exposed--api-key,--api-base,--dimensions, and--syncflags for LiteLLM-specific configurationTextProjectorCallbacktype and provider-specific implementations (_project_text_with_sentence_transformers,_project_text_with_litellm) so that we can keep benefitting from the existing caching approach regardless of the model used to compute the projectionsTesting
Verified in packages/backend/examples/notebook.ipynb with:
nomic-embed-text)text-embedding-3-small)