pgvector is a PostgreSQL extension that adds vector similarity search capabilities to Postgres databases. It allows developers to store and query vector embeddings directly in PostgreSQL, making it ideal for AI and machine learning applications that need semantic search, recommendation systems, or similarity matching. The extension supports multiple distance functions like L2, cosine, and inner product, and offers indexing options including HNSW and IVFFlat for fast approximate nearest neighbor search. It integrates seamlessly with existing Postgres infrastructure and can be used from any programming language with a Postgres client.
Added on
