#vector-search

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

The SQLite of Vector Search & Episodic Memory for AI Agents in ~120KB. Pure C99, AVX2+FMA, ARM NEON, FASM x64, Native MCP Server.

Ruby wrapper for the Weaviate vector search database API

Python SDK for enVector - encrypted vector search powered by homomorphic encryption

AI-powered persistent memory system

Local-first markdown context engine for AI coding agents. Indexes a repo's docs into SQLite (FTS5 + local vectors) and exposes CLI, HTTP, and MCP over one service core. Ships a docs registry, skills/agents sync, and code-to-doc governance.

Embedded PostgreSQL with pgvector extension for Node.js applications

Go toolkit for document ingestion and retrieval.

Vector memory that gets smarter every time your agent uses it.

Go SDK for BarqDB — an embedded local-first database with HNSW vector search built in.

Experimental Rust vector-compression sidecars implementing TurboQuant (ICLR 2026), PolarQuant and QJL research primitives, with exact reranking and workload-scoped benchmark receipts.

Local hybrid (vector + BM25) code search for AI coding agents — AST chunking, MiniLM embeddings on ONNX, Go call graph, MCP server. No external services.

Lightweight QMD alternative: fully local hybrid search engine (BM25 + vector + rerank) and MCP server in a single ~15MB Go binary. No Node.js. Runs on Raspberry Pi.

Embedding search for Go: the whole index scan is one matrix-vector product. 18-38x a hand-written loop. No cgo.

A high-performance SWAR-accelerated lexer for the LibraVDB query language — SQL, graph patterns, and vector similarity. Provided by LibraVDB INC.