whereismyai
Pg Vectorsearch
Pg Vectorsearch is a PostgreSQL extension for high-performance approximate nearest neighbor search using PRISM indexes.
Pg Vectorsearch is a PostgreSQL extension designed for high-performance approximate nearest neighbor search. It introduces the PRISM index, an access method built around quantization, routing, and partitioning techniques. The extension uses RaBitQ quantization and SIMD-optimized distance computation to accelerate similarity matching on large datasets. It supports native vector types as well as existing pgvector tables without requiring query changes. The software is written in C and targets PostgreSQL users who require scalable vector operations.
Pg Vectorsearch at a glance
Key features
- PRISM index for IVF-style routing
- RaBitQ quantization with error bounds
- SIMD-optimized distance computation
- Dynamic updates with posting-list splits
- Pgvector compatible table indexing
- L2, inner product, and cosine distance support
Use it for
- Approximate nearest neighbor search in PostgreSQL
- Scaling vector similarity queries on large datasets
- Indexing existing pgvector columns without query changes
Alternatives to Pg Vectorsearch
- zvec — It is a pure local, serverless vector database providing fast hybrid search.
- Glider — It combines crash-safe S3 durability with fast local SSD caching for vector search.
- seekdb — Unifies relational, vector, and text data for production-grade AI agents.
- LEANN — Efficiently index millions of documents locally with 97% storage savings.
- LightRAG — It uses a dual-layer architecture to reduce LLM calls and retrieval latency.
- PageIndex — It provides 98.7% accuracy on FinanceBench without vector databases.
- UltraRAG — Build complex RAG workflows using modular MCP components and YAML configuration.
- Pagesift — Extracts web data with high memory efficiency and speed for AI agents.
- Vizro — Combines low-code configuration with high-code customization for data dashboards.
- Leviathan — It indexes records into a single SQLite file for fast, cited agent queries.
- Morphik — Provides native Model Context Protocol support and granular multi-user access.
- Finetuning With Sampling — Official implementation of the Finetuning with Sampling research paper from Harvard.
All Pg Vectorsearch alternatives