Awesome Vector Databases

A curated registry of vector databases, vector stores, vector-search libraries, and benchmarks for AI, RAG, and semantic-search applications.

The goal is to be the definitive place to discover and compare production-grade vector data systems — from managed cloud services and self-hosted engines to embedded libraries and hybrid search platforms.

Inclusion criteria: Resources must provide native vector storage, indexing, or similarity search as a core capability. General SQL databases without vector extensions are out of scope. See CONTRIBUTING.md for the full quality bar.

---

Contents

• Managed Vector Databases
• Self-Hosted Vector Databases
• Embedded Vector Stores
• Graph + Vector
• Time-Series + Vector
• Vector Search Libraries
• Hybrid Search
• Benchmarks
• Related Awesome Lists
• Contributing
• License

---

Managed Vector Databases

Fully managed cloud services that provide vector storage, indexing, and similarity search without infrastructure overhead.

• Pinecone Official — Managed vector database with metadata filtering, hybrid search, and no-index tuning.
• Zilliz Cloud Official — Fully managed Milvus service with GPU indexing and enterprise security.
• Weaviate Cloud Official — Managed vector search engine with built-in vectorization and GraphQL interface.
• Qdrant Cloud Official — Managed Qdrant cluster service with quantization and hybrid search.
• DataStax Astra DB Official — Managed Cassandra-compatible database with vector search for RAG.
• Chroma Cloud Official — Managed service for the Chroma embedded vector store.
• Azure AI Search Official — Cloud search service with vector and hybrid retrieval.
• AlloyDB AI Official — Google Cloud managed PostgreSQL with vector search and AI integrations.
• SingleStore Official — Distributed SQL database with SingleStore-V integrated vector search.
• MyScale Official — ClickHouse-based managed vector database for structured + vector data.

---

Self-Hosted Vector Databases

Production-ready vector databases you can deploy on your own infrastructure.

• Milvus Official — Open-source distributed vector database with GPU support and cloud-native architecture.
• Qdrant Official — Open-source vector database in Rust with filtering, quantization, and hybrid search.
• Weaviate Official — Open-source vector search engine with modular AI integrations.
• pgvector Official — PostgreSQL extension for vector similarity search with HNSW and IVFFlat indexes.
  • Install: CREATE EXTENSION vector; or use the pre-built Docker image.
• Chroma Official — AI-native open-source embedding database for local and production deployments.
• Vespa Official — Big data serving engine with native vector search, ranking, and tensors.
• Vald Official — Cloud-native, highly scalable distributed vector search engine.
• Marqo Official — End-to-end vector search engine with built-in embedding generation.
• OpenSearch Official — Open-source search and analytics suite with k-NN vector search.
• Elasticsearch Official — Distributed search engine with dense and sparse vector search.
• Redis Stack Official — In-memory data platform with vector similarity search via RediSearch.
• Couchbase Official — Distributed NoSQL database with vector search for RAG applications.

---

Embedded Vector Stores

Lightweight, in-process vector stores and libraries for local development and edge deployments.

• Chroma Official — Popular embedded vector store with simple Python/JS APIs.
• FAISS Official — Facebook AI Similarity Search library for efficient nearest-neighbor search.
• LanceDB Official — Serverless vector database built on the Lance columnar format.
• SQLite-vec Community — SQLite extension for vector search, designed for edge and embedded use.
• Voyager Official — Spotify's approximate nearest-neighbor search library with a Python/Java/C++ API.
• HNSWLib Community — Header-only HNSW implementation with Python bindings.
• Annoy Official — Spotify's approximate nearest-neighbor library using random projections.
• ScaNN Official — Google's scalable nearest-neighbor search library.
• USearch Official — Fast, compact, and broadly compatible single-file similarity search engine.
• Vectorlite Community — SQLite-like embedded vector database with HNSW indexing.

---

Graph + Vector

Graph databases that combine vector similarity search with relationship traversal for GraphRAG and knowledge-graph workloads.

• Neo4j Official — Graph database with native vector indexes and GenAI integrations.
• ArangoDB Official — Multi-model database with ArangoSearch and vector search for documents and graphs.
• TigerGraph Official — Native parallel graph database with TigerVector hybrid graph+vector search.
• Memgraph Official — In-memory graph database with native vector search and graph traversal.
• Kùzu Official — Embedded property graph database with built-in vector and full-text search.
• FalkorDB Official — Low-latency graph database with vector support for RAG.
• Amazon Neptune Official — Managed graph database with vector search and Neptune Analytics.

---

Time-Series + Vector

Databases that combine time-series/analytics workloads with vector search, often via SQL extensions.

• TimescaleDB Official — Time-series PostgreSQL extension with pgvector and pgvectorscale for time-aware hybrid search.
• ClickHouse Official — Real-time OLAP database with vector distance functions and ANN indexing.
• DuckDB Official — In-process analytical database with array distance functions and vector-search extensions.
• StarRocks Official — Fast OLAP engine with native HNSW/IVFPQ vector indexes and hybrid queries.
• Apache Doris Official — Real-time OLAP data warehouse with vector index support for high-dimensional data.

---

Vector Search Libraries

Low-level libraries and algorithms for building approximate nearest-neighbor indexes.

• FAISS Official — Comprehensive library for efficient similarity search and clustering of dense vectors.
• HNSWLib Community — Fast HNSW graph-based nearest-neighbor library.
• Annoy Official — Approximate nearest-neighbor library optimized for memory-mapped indexes.
• ScaNN Official — Google's library for vector similarity search at scale.
• USearch Official — Single-file similarity search library compatible with many index formats.
• NMSLIB Community — Similarity search library and toolkit with non-metric space support.
• DiskANN Official — Disk-resident ANN index from Microsoft for billion-scale datasets.
• Vearch Official — Scalable vector search system built on Faiss for deep-learning retrieval.

---

Hybrid Search

Search engines that combine keyword/BM25, filtering, and vector similarity in a single platform.

• Meilisearch Official — Instant search engine with vector search, typo tolerance, and faceting.
• Typesense Official — Open-source typo-tolerant search engine with vector and semantic search.
• Algolia Official — Managed search platform with hybrid neural and keyword search.
• Manticore Search Official — Open-source search engine with SQL-like queries and vector search.
• Quickwit Official — Cloud-native search engine for logs and traces with vector search support.
• Apache Solr Official — Open-source search platform with dense vector search via the DenseVector field type.

---

Benchmarks

Tools and datasets for measuring vector database and embedding performance.

• VectorDBBench Official — Open-source benchmark tool for vector databases and cloud services.
• ANN-Benchmarks Community — Standardized benchmarking suite for approximate nearest-neighbor algorithms.
• Big-ANN Benchmarks Official — NeurIPS competition benchmark for billion-scale ANN search.
• BEIR Community — Heterogeneous benchmark for information retrieval with zero-shot evaluation.
• MTEB Community — Massive Text Embedding Benchmark leaderboard for embedding models.
• VIBE Community — Vector Index Benchmark for Embeddings using modern embedding datasets.

---

Related Awesome Lists

Discover complementary resources for building RAG, agent, and local-LLM systems.

• Awesome RAG — Curated list of open-source tools for retrieval-augmented generation.
• Awesome MCP Servers — Curated list of Model Context Protocol servers, including vector-store integrations.
• Awesome Local LLM — Curated list of tools and models for running LLMs locally.

---

Contributing

Read CONTRIBUTING.md for the quality bar, entry format, and PR process.

---

License

This list is released into the public domain under CC0-1.0.

Want us to build this for you?

Enterprise AI Atlas is maintained by Vibe Coding Agency. We prototype and ship agentic systems, MCP servers, and enterprise AI integrations for teams that need working software fast — without hiring a full AI engineering team.

Free guide: The Non-Technical Founder's Guide to Agentic AI — what agents and MCP servers are, and how to get a system built.