langchain - 💡(How to fix) Fix Add generic Hybrid Retriever with BM25 + FAISS + RRF Fusion [2 comments, 2 participants]

Submission checklist

• This is a feature request, not a bug report or usage question.
• I added a clear and descriptive title that summarizes the feature request.
• I used the GitHub search to find a similar feature request and didn't find it.
• I checked the LangChain documentation and API reference to see if this feature already exists.
• This is not related to the langchain-community package.

Package (Required)

• langchain
• langchain-openai
• langchain-anthropic
• langchain-classic
• langchain-core
• langchain-model-profiles
• langchain-tests
• langchain-text-splitters
• langchain-chroma
• langchain-deepseek
• langchain-exa
• langchain-fireworks
• langchain-groq
• langchain-huggingface
• langchain-mistralai
• langchain-nomic
• langchain-ollama
• langchain-openrouter
• langchain-perplexity
• langchain-qdrant
• langchain-xai
• Other / not sure / general

Feature Description

Add a generic, vendor-independent HybridRetriever that combines BM25 sparse retrieval and FAISS dense vector search using Reciprocal Rank Fusion (RRF).

Current hybrid search examples in LangChain are tied to specific vendors (Pinecone, Weaviate). There is no standalone generic implementation using RRF.

Proposed class: HybridRetriever

• BM25 search (rank-bm25)
• FAISS vector search (cosine similarity)
• RRF fusion for combining results
• Optional cross-encoder reranking

Use Case

Developers building RAG systems often need hybrid retrieval without being locked into a specific vector database. A generic HybridRetriever with RRF would work with any document list and serve as a strong baseline for production RAG pipelines.

Proposed Solution

A standalone HybridRetriever class with:

• bm25_search()
• vector_search()
• rrf_fusion()
• retrieve()

Working implementation available here: [tera GitHub repo link daalo]

Alternatives Considered

Existing vendor-specific implementations (pinecone_hybrid_search.py, weaviate_hybrid_search.py) but these require external services.

Additional Context

No response