pgvector
Vector similarity search as a Postgres extension
pgvector is profiled here as a Vector Database tool for engineering teams. Read about features, pricing, and how it compares to related options in the tools directory.
Description
pgvector is an open-source extension that adds vector storage and similarity search to PostgreSQL, created by Andrew Kane. It introduces a vector column type with distance operators, so embeddings live in the same database as relational data and join directly to it in plain SQL. Teams that already run Postgres adopt it to add semantic search without standing up separate vector infrastructure, and most managed Postgres providers support it. Keeping vectors beside relational data simplifies architecture, which has made the extension a popular first step into semantic search. Because it ships with most managed Postgres services, teams add it without changing databases.
Key Capabilities:
Vector column type with L2, inner product, and cosine distance
HNSW and IVFFlat indexes for approximate nearest-neighbor search
Exact nearest-neighbor search for smaller datasets
SQL joins between embeddings and relational data
Half-precision and binary vectors for storage efficiency
PostgreSQL license with broad managed-provider support
Alternative tools
- Deep Lake
Database for AI that stores tensors and embeddings
- Marqo
Open-source vector search with built-in embedding inference
- Chalk
Feature platform for real-time machine learning data
- Hopsworks
Feature store and ML platform for batch and real-time data
- turbopuffer
Serverless vector and full-text search on object storage
- LanceDB
Embedded multimodal vector database on the Lance format