vector-database

Mintplex-Labs/anything-llm

A local-first LLM workspace that turns documents into working agents — AnythingLLM bundles RAG chat, multi-user support, and configurable AI agents into one self-hosted app, aimed at people who want to run document-grounded chat and agentic workflows locally instead of relying on hosted AI services. It targets self-hosters and teams needing private, multimodal, agent-capable LLM tooling without complex setup.

milvus-io/milvus

A scalable vector database built to back LLM and RAG pipelines — Milvus is a cloud-native vector database written in Go and C++ for large-scale ANN search across text, image, and multi-modal embeddings. It supports multiple index types like HNSW and DiskANN, and suits teams building LLM, RAG, or similarity-search applications needing distributed, production-grade scale.

volcengine/OpenViking

A unified, self-evolving context store for memory, RAG and skills — OpenViking is Volcengine's open-source context database that merges agent memory, knowledge RAG, and reusable skills into one self-evolving store. It targets developers building AI agents who need a single backend instead of stitching together separate vector DBs, memory layers, and plugin systems.

VectifyAI/PageIndex

RAG without vectors: PageIndex reasons over document trees instead — PageIndex replaces vector DBs and chunking with a tree-structured index that an LLM reasons over, mimicking how a human expert navigates a document. New SDK supports local and cloud modes, plus a fast 'Flash' indexer and a File System layer for scaling across millions of documents. Built for developers building accurate RAG over long, complex professional documents.

qdrant/qdrant

Rust-built vector search engine for production AI workloads — Qdrant is a Rust-based vector database with extended filtering support, built for semantic search, recommendations, and neural-network matching at scale. It offers a production-ready API and a managed cloud option, aimed at developers building AI apps that need fast, reliable similarity search under heavy load.

topoteretes/cognee

Persistent long-term memory for AI agents, free and open source — Cognee gives AI agents long-term memory using small models, combining vector and graph databases for context engineering instead of fine-tuning. It targets developers building agents that need to recall knowledge across sessions, with community plugins extending its core.

xerj-org/xerj

A single Rust binary that gives AI coding agents real search — XERJ indexes code, docs, logs and PDFs locally for BM25, vector and hybrid search behind an Elasticsearch-compatible API. Its reference-coding feature lets agents like Claude Code or Cursor find and reuse solutions from similar open-source repos instead of re-reading files, cutting output tokens by up to 2.7x in testing.

memvid/memvid

A Single-File Memory Layer for AI Agents, No Database Needed — Memvid stores embeddings and knowledge in one portable, versioned file instead of a database, letting agents get instant retrieval and long-term memory offline. Rewritten in Rust with Python bindings, it targets teams wanting to drop complex RAG pipelines for something simpler and file-based.

neuml/txtai

A Single Framework Combining Vector Search, Graphs and LLM Agents — txtai bundles vector indexes, graph networks and relational databases into one embeddings database that powers semantic search, RAG and multi-model workflows. Rather than stitching together separate vector DB, graph and LLM tools, developers get a unified foundation for building autonomous agents and search-driven AI applications in Python.

HelixDB/helix-db

A Rust graph database that bundles vector and full-text search — HelixDB combines a labeled property graph, approximate nearest-neighbor vector search, and BM25 full-text search in one transactional Rust engine on object storage. A CLI spins up local instances in Docker or embeds directly into an app, with queries callable from Rust, TypeScript, Go, or Python—useful for RAG and AI apps needing graph plus vector in one store.

kividbio/kividb

A Rust-built Redis alternative that claims 9x the throughput — KiviDB is a multi-threaded, Redis-protocol-compatible server written in Rust with built-in full-text and vector search. Its benchmarks claim 21M pipelined GETs/sec on 32 cores—about 9x Redis 8 and 2.3x Dragonfly, with lower p99 latency. Useful for teams wanting a drop-in Redis/Valkey replacement with search features baked in.