pgvector
Vector similarity search extension for PostgreSQL
View on GitHubWhat it solves
Infrastructure overhead for teams already on Postgres: semantic search without adding a separate vector database to operate.
Overview
pgvector adds vector storage and similarity search to PostgreSQL, exact and approximate nearest-neighbor search using L2 distance, inner product, and cosine distance. If you already run Postgres, pgvector lets you store embeddings alongside the rest of your data with no additional infrastructure.
Key facts
- Language
- C
- License
- PostgreSQL
- Maturity
- Stable
- Maintainer
- pgvector
- Reviewed
- 2026-06-11
Where to start
The indexing section explains the HNSW and IVFFlat trade-offs that determine query performance at scale. Read this before choosing an index type.
README.mdTechnologies
More in this domain
ChromaDB
Open-source embedding database for AI applications
Persisting, querying, and filtering vectors without standing up a full database stack.
LlamaIndex
Data framework for connecting LLMs to custom data sources
Turning heterogeneous data sources into LLM-queryable indices without writing custom parsers and retrieval pipelines for each format.