Awesome Vector Search (Open-Source, 2024+)

Curated list of open-source vector-native databases, databases with vector column support, vector search & indexing libraries, cloud services, benchmarks, and research papers.
All GitHub links for databases & libraries; lists ordered by fork count (descending).

---

Vector-Native Databases

Purpose-built systems where vector search is a first-class, core workload.

Name (GitHub)	Stars	Forks	Last Commit	Status	Description
Milvus	~43k	~4k	2026-01	Active	Distributed, cloud-native vector database
Qdrant	~29k	~2k	2025-10	Active	Rust-powered vector DB with hybrid filtering
ChromaDB	~26k	~2k	2026-01	Active	Embedding-first vector store for LLM/RAG workloads
Endee.io	~1k	~1.1k	2026-03	Active	cost-efficient, high-throughput vector search - scaling to 1B+ vectors per node
Weaviate	~16k	~1k	2026-01	Active	Schema-aware vector DB with GraphQL
LanceDB	~9k	~800	2026-01	Active	Data-lake-native vector database built on Apache Arrow
Vespa	~7k	~700	2026-02	Active	Large-scale search engine with native vector support
Vald	~2k	~90	2026-01	Active	Kubernetes-native distributed vector search

---

Databases with Vector Column / Extension Support

General-purpose databases that add vector similarity search as a feature or extension.

Name (GitHub)	Stars	Forks	Last Commit	Status	Description
ClickHouse	~46k	~8k	2026-01	Active	Columnar analytical DB with native vector search functions
DuckDB	~36k	~3k	2026-01	Active	Analytical DB with integrated vector similarity search
pgvector	~20k	~1k	2026-01	Active	PostgreSQL extension for vector similarity
Redisearch	~6k	~570	2025-11	Active	Redis module supporting ANN vector search
pgvectorscale	~3k	~120	2025-11	Active	Production-grade scaling & indexing layer for pgvector

---

Vector Search & Indexing Libraries (ordered by forks)

Name (GitHub)	Stars	Forks	One-liner
FAISS	~39k	~4k	High-performance similarity search (CPU/GPU)
Annoy	~14k	~1k	Tree-based ANN optimized for read-heavy workloads
hnswlib	~7k	~1.3k	Header-only HNSW ANN index
ScaNN	~3k	~600	Google’s scalable ANN library
NMSLIB	~5k	~470	Flexible ANN search library
DiskANN	~2k	~400	Disk-based ANN for billion-scale vectors
USearch	~4k	~300	SIMD-optimized vector search
NGT	~1.5k	~250	Graph-based ANN index
Autofaiss	~1k	~80	Automatic FAISS index configuration

---

Cloud Services (Open-Source Core / Hosted)

Managed offerings built on open-source vector engines.

• Qdrant Cloud – Managed Qdrant with scaling & backups
• Zilliz Cloud – Managed Milvus service
• Weaviate Cloud – Hosted Weaviate with enterprise integrations
• LanceDB Cloud – Managed LanceDB optimized for analytics
• Endee Cloud – Managed Endee.io vector database service

---

Benchmarking Tools

Name (GitHub)	Description
ANN-Benchmarks	De-facto standard ANN algorithm benchmark
VectorDBBench	Benchmark framework for vector databases
VIBE-Benchmark	Benchmark suite for vector index quality & speed
annlite-benchmark	Lightweight ANN benchmarking suite

---

Research Papers & Links

Title	Link
Efficient and Robust Approximate Nearest Neighbor Search Using HNSW	https://arxiv.org/abs/1603.09320
FAISS: A Library for Efficient Similarity Search and Clustering	https://arxiv.org/abs/1702.08734
ScaNN: Efficient Vector Similarity Search at Scale	https://arxiv.org/abs/1908.10396
DiskANN: Fast & Accurate Billion-Point NN Search	https://arxiv.org/abs/2011.01608
Product Quantization for Nearest Neighbor Search	https://hal.inria.fr/inria-00514496
VIBE: Vector Index Benchmark for Embeddings (2025)	https://arxiv.org/abs/2505.17810
MicroNN: On-Device Disk-Resident Updatable Vectors (2025)	https://arxiv.org/abs/2504.05573

---

Notes

• Open-source only
• Actively maintained in 2024–2026
• Fork counts are approximate and used for relative ordering