Open SourceRAG17k
pgvector
Postgres extension adding vector columns and ANN indexes.
pgvector is an open-source extension that adds vector data types and similarity search to PostgreSQL. It supports exact and approximate nearest-neighbor search using IVFFlat and HNSW indexes with L2, inner-product, and cosine distance, all queryable through standard SQL. It is notable for letting teams add vector search to an existing Postgres database rather than running a separate vector store.
Repository
pgvector/pgvectorLanguage
CWhat you'd build with it
- Adding semantic search and RAG retrieval to an app that already uses PostgreSQL, avoiding a new datastore
- Combining vector similarity with relational joins and SQL filters in a single query
- Storing embeddings alongside transactional data with the same backup, consistency, and access controls
Tags
postgresvectorextension