Netics has released pg_pinecone, an open-source PostgreSQL extension that integrates vector search capabilities directly into the database. This extension offers three engines: a native engine that stores compressed vector codes within PostgreSQL, a pgvector engine that interfaces with existing HNSW indexes, and a remote engine that directs searches to Pinecone's managed service. Benchmarks suggest the native engine provides significantly smaller indexes and faster build times compared to traditional HNSW implementations, while maintaining high recall rates, especially when combined with filtering. AI
IMPACT Enables developers to integrate vector search directly into their existing PostgreSQL databases, simplifying AI application development.
RANK_REASON This is a new open-source extension for an existing database system, offering new functionality but not a core AI model release or significant industry shift.
- Anthony Mahanna
- Apache Software License 2.0
- GitHub
- Harshita Daddala
- Hierarchical Navigable Small-World graphs
- Netics
- pg_pinecone
- pgvector
- Pinecone
- PostgreSQL
- RaBitQ: Quantizing High-Dimensional Vectors with a Theoretical Error Bound for Approximate Nearest Neighbor Search
- SQL
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →