pgvector
Vector search for PostgreSQL
About pgvector
pgvector is an open-source extension that adds vector similarity search capabilities to PostgreSQL. It enables developers to store and query embeddings directly within a relational database, allowing seamless integration with existing applications and workflows. pgvector supports efficient nearest neighbor search and is commonly used in applications where structured relational data and vector search need to coexist. By extending PostgreSQL rather than introducing a separate database, pgvector simplifies infrastructure while enabling AI-powered search and recommendation functionality.
Key features
Pricing
Common use cases
Common questions about pgvector
Is pgvector free to use?
Yes, pgvector is open source under the PostgreSQL License license. You can self-host it at no licensing cost, though you'll need to manage infrastructure and operational costs.
Can I get support for pgvector?
pgvector has community support through documentation, forums, and GitHub issues. Some open-source databases also offer commercial enterprise support contracts.
What are the main use cases for pgvector?
pgvector is commonly used for adding vectors to existing apps, hybrid relational + vector, prototyping, and similar applications requiring semantic similarity search.
Does pgvector integrate with popular AI tools?
Most vector databases integrate with LangChain, LlamaIndex, and popular embedding providers. Check the pgvector documentation for specific integration guides and examples.
Comparisons featuring pgvector
Not sure if pgvector is right for you?
Compare it side-by-side with other vector databases to find the best fit for your project.