Tool Open source
pgvector is an open-source PostgreSQL extension for storing vectors alongside relational data and querying exact or approximate nearest neighbors. It supports single-precision, half-precision, binary, and sparse vectors, with L2, inner-product, cosine, L1, Hamming, and Jaccard distance operators.
The extension uses standard PostgreSQL tables and queries, including inserts, updates, bulk loading with COPY, filtering, joins, and indexed nearest-neighbor searches. It retains PostgreSQL capabilities such as ACID compliance and point-in-time recovery, and provides quantization options for scaling vector collections.
It supports PostgreSQL 13 and later and can be compiled on Linux, macOS, and Windows or installed through Docker, package managers, PGXN, and hosted PostgreSQL providers.
1 use taken from transcripts — each links to the moment in the video.
PostgreSQL extension used to store embeddings and support similarity lookups.
2 in the library.