About pgvector
An open-source PostgreSQL extension that adds a vector column type and similarity search operators to Postgres, enabling embedding storage and ANN/ENN queries within the database.
Key Features
- Vector column type, indexing, and distance operators (L2, inner product, cosine, L1) inside Postgres with ACID and replication support.
- Supports exact and approximate nearest-neighbor searches, multiple numeric/storage formats (FP32, FP16, binary, sparse).
- Works with any Postgres client and preserves Postgres features like JOINs, transactions, and backups.
Use Cases & Best For
Developers who want to store vectors alongside relational data and leverage Postgres features (transactions, joins, replication).
Teams preferring to add vector search to existing Postgres deployments without a separate vector store.
About Vector Databases
Storage and retrieval for embeddings