Pg_vectorsearch: An index that is faster than pgvector's IVFFlat and HNSW
Developers introduced pg_vectorsearch, a PostgreSQL extension that provides a new indexing method for similarity search on high‑dimensional vectors. Benchmarks show it outperforms the existing pgvector approaches, IVFFlat and HNSW, delivering lower query latency and reduced index size. The project is open‑source and aims to simplify integration of fast vector search for modern AI applications within relational databases.