pgvector

pgvector is a PostgreSQL extension that adds vector similarity search capabilities to PostgreSQL, enabling it to function as a vector database within your existing relational database. It supports storing embeddings as vector columns, creating HNSW or IVFFlat indexes for approximate nearest neighbor search, and filtering by metadata using standard SQL WHERE clauses. pgvector is popular for small to medium RAG deployments because it eliminates the need for a separate vector database — you can store both application data and embeddings in one place. It supports cosine distance, L2 distance, and inner product similarity metrics.