bm25

Open-source, S3-native vector and full-text search engine. Fast, cheap, self-hostable.

Semantic code search plugin for Claude Code using hybrid vector search + BM25. 98% accuracy, 5x faster than grep.

Rust-based retrieval system with hybrid search (vector + BM25), async ingestion, and gRPC-first API

Contextual Retrieval solves this problem by prepending chunk-specific explanatory context to each chunk before embedding (“Contextual Embeddings”) and creating the BM25 index (“Contextual BM25”).

sovereign embedded vector database, single-file .nest container with content-addressable citations, reproducible builds, and offline-first model verification.

Lightning-fast RAG for AI agents. ONNX-powered, 4-layer fusion, MCP server. No PyTorch.