pgvector
PostgreSQL extension that adds a native vector column type and indexing for similarity search in a relational database (ANN search inside Postgres).
vector-database-frameworksNew to PyRadar
64
Hero Score
Popularity
72
Performance
70
Ecosystem
50
Maturity
69
Dev Experience
57
⭐ 21,548 stars⬇ 5.7M downloads/wkFirst release: Jun 2021Last release: Feb 2026
Async Support: YesPlugin Extensions: MediumSpeed: MediumDoc Quality: HighLearning Curve: Medium
Pros
- • Adds vector search directly into PostgreSQL — no separate system required
- • Leverages Postgres tooling (ACID, joins, security, backup/restore)
- • Works with IVFFlat/HNSW index types for adjustable performance
Cons
- • PostgreSQL is not optimized as a standalone vector engine at extreme scale
- • Queries involving heavy vector ops can slow regular SQL workloads
- • Not ideal for extremely low-latency or billion-vector scale without sharding