# pgvector

Open-source vector similarity search for Postgres

Repository: https://github.com/pgvector/pgvector
Canonical: https://ross.abutalabs.com/products/pgvector
Language: C
License: NOASSERTION
License Family: other
Topics: nearest-neighbor-search, approximate-nearest-neighbor-search
Last push: 2026-08-20T03:47:50+00:00

## Health v2 (maintenance only)
Score: 76/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 98, release rhythm 35, longevity 100
- inputs: {"age_days": 1961, "days_push": 13, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 22760, forks 1297 (observed 2026-08-28T04:11:34.000032+00:00)

## What it is
pgvector is an open-source PostgreSQL extension that adds vector data types and vector similarity search, supporting exact and approximate nearest neighbor search with multiple distance metrics. It lets you store embeddings alongside relational data while retaining full Postgres features like ACID compliance and JOINs.

## Use cases
- add vector similarity search to postgres
- store embeddings with relational data
- build a rag application without a separate vector database
- find nearest neighbors by cosine distance in sql
- semantic search over documents in postgres
- recommendation system with embedding similarity

## When to choose
- you already use Postgres and want vector search without running a separate database
- you need ACID transactions, JOINs, and backups for your vector data
- you want a mature, widely supported extension with clients in many languages

## When to avoid
- you need massive-scale vector search beyond what a single Postgres instance can handle
- you require advanced vector database features like multi-tenancy filtering at extreme scale or built-in hybrid search
- you don't use Postgres and have no reason to adopt it

## Facets
- artifact type: library
- maturity: active
- function: vector-database, search-engine, database
- domain: databases, machine-learning
- platform: windows, self-hosted, cloud
- tags: postgres-extension, vector-similarity-search, embeddings, nearest-neighbor-search, hnsw, ivfflat, sql, search, retrieval-augmented-generation, linux, macos, docker

## Member repositories
- pgvector/pgvector (main) score 76
- pgvector/pgvector-python (sdk) score 73

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:34.000032+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T16:56:55.905354+00:00, confidence not recorded.
  - readme: https://github.com/pgvector/pgvector (fetched 2026-08-28T04:11:34.000032+00:00, sha 305c01a02eff)
- Data as of 2026-08-30T08:39:29.467469+00:00.