Vector databases · pgvector project (open-source, community-maintained)

pgvector

Open-source PostgreSQL extension that adds vector similarity search directly inside a Postgres database.

Visit pgvector Alternatives

pgvector is an open-source extension that adds a vector data type and similarity search operators to PostgreSQL, rather than being a standalone database of its own. It lets teams store embeddings in the same database as their relational data and run exact or approximate nearest-neighbor search (via IVFFlat or HNSW indexes) alongside ordinary SQL joins, transactions and filtering, avoiding the need to sync data into a separate vector store. Because it is a Postgres extension, it inherits Postgres's durability, backup, replication and access-control model, and it runs anywhere Postgres runs, including most managed Postgres services that support installing extensions. Its main trade-off against purpose-built vector databases is scale and specialized indexing performance: pgvector is generally favored by teams with modest vector workloads who value keeping everything in one relational database over teams needing billion-scale, low-latency vector search.

At a glance

Vendor pgvector project (open-source, community-maintained)
Pricing model Open source + paid options
Free tier Yes
Deployment Self-hosted
Open source Yes (PostgreSQL License)
Best for Teams already on PostgreSQL who need vector search without adding a separate database system.

Pricing

Free and open source; there is no separate vendor or pricing since it is an extension run inside a user's own PostgreSQL instance.

Pricing has not been verified yet — see the vendor's site.

Features

  • Vector data type and distance operators inside PostgreSQL
  • Approximate nearest-neighbor indexes (IVFFlat, HNSW)
  • Combines vector search with standard SQL joins and filtering
  • Inherits Postgres transactions, backups and replication
  • Runs on self-hosted Postgres or any managed Postgres that allows extensions
  • No separate infrastructure or service to operate

Integrations

Profile last reviewed September 21, 2026

Alternatives

Chroma Open-source embedding database designed for simple local-first development, with a managed serverless cloud tier. Qdrant Open-source vector database written in Rust, focused on filtered similarity search with a free managed cloud tier. Milvus Open-source vector database built for billion-scale similarity search, governed under the Linux Foundation.

pgvector in the index now

Terms to know

Machine learning Building models that learn patterns from data to make predictions or decisions, rather than following hand-coded rules. Semantic search Search that matches results to a query's meaning rather than its exact keywords, using vector similarity between embeddings. Vector database A database optimized to store numeric embeddings and quickly find the ones most similar to a given query vector.

Related guides

How to choose a vector database Pick a vector database by deployment model, scale and search type, not by benchmark charts — most teams' real constraint is who runs the cluster.