Vector database

A store optimised for indexing and querying high-dimensional vectors, used to find semantically similar items for RAG and other retrieval tasks.

The core operation is approximate nearest neighbour search, trading recall for speed. The choice between a dedicated vector store and an extension to an existing database (pgvector) is a decision about operational complexity versus query capability.

Explained in depth

pgvector

Vector similarity search in PostgreSQL - where the data physically lives across heap, TOAST and index, how HNSW and IVFFlat retrieve it, and how pgvector serves as the memory layer in a RAG or agent system.

Implemented by

pgvector
Dedicated vector store

Related concepts

Vector search
Embedding
Retrieval-augmented generation