Vector search in PostgreSQL. The foundation for AI assistants and knowledge bases that understand your own data.
- Type
- PostgreSQL extension
- Function
- Vector search
- Role
- RAG & AI search
What is PgVector?
PgVector is an extension for PostgreSQL that lets you store and search embeddings. Embeddings are numerical representations of text or images. They let you search by meaning instead of exact words. This is the foundation of RAG: AI that answers based on your documents.
Why PgVector?
Search by meaning
Find relevant information, even when different words are used.
In your existing database
No separate vector database needed, everything in PostgreSQL.
Foundation for RAG
AI assistants that answer with sources from your knowledge base.
What we use it for
- AAI knowledge bases and internal assistants
- BSmart search in portals
- CProduct recommendations
- DMatching and classifying documents
When do you choose PgVector?
Choose PgVector if you want AI to work with your own documents and data, without managing an extra database.
PgVector or a separate vector database?
For most applications, a separate vector database is unnecessary. With PgVector, embeddings sit alongside your regular data, with the same permissions, backups and transactions. That is simpler and safer.