We’re excited to announce the release of our pg_embedding extension for Postgres and LangChain! The new pg_embedding extension brings 20x the speed for 99% accuracy to graph-based approximate nearest neighbor search to your Postgres databases.  While the pgvector extension with IVFFlat indexing has been a popular choice, our new pg_embedding extension uses Hierarchical Navigable Small […]

Neon

The pg_embedding extension for Postgres and LangChain introduces a 20x speed improvement for high-dimensional vector similarity searches using the HNSW index. It offers advantages over the IVFFlat indexing used by pgvector, particularly in terms of search speed and accuracy. Benchmark tests show that HNSW provides faster performance with similar or better accuracy compared to IVFFlat. The choice between pg_embedding and pgvector depends on specific use case requirements, such as memory constraints and desired search speed.

20x faster than pgvector: introducing pg_embedding extension for vector search in Postgres and LangChain