pgvecto.rs
PostgreSQL extension for scalable, low-latency vector search written in Rust. Features 20x faster HNSW than pgvector, with support for FP16, INT8, and binary vectors. This is an OSS extension.
About this tool
Overview
pgvecto.rs is a Postgres extension written in Rust that provides scalable, low-latency, hybrid-enabled vector search. Built on the pgrx framework, it offers significant performance improvements over pgvector while maintaining PostgreSQL compatibility.
Key Features
- High Performance: HNSW algorithm 20x faster than pgvector at 90% recall
- Multiple Data Types: FP16 (16-bit float), INT8 (8-bit integer), binary vectors
- WAL Support: Write-Ahead Logging for index and data durability
- Filtering Support: Advanced filtering during search and retrieval
- Memory Safety: Rust's compile-time checks guarantee memory safety
- Hybrid Search: Combines vector and traditional search capabilities
Why Rust?
Rust provides:
- Memory safety without garbage collection
- Prevention of entire classes of bugs and security issues
- Better performance than C extensions
- Modern development experience with cargo ecosystem
Performance Comparison
At 90% recall:
- 20x faster than pgvector HNSW
- Lower latency for production workloads
- Better resource utilization
Data Type Support
- FP32: Standard 32-bit floating point
- FP16: 16-bit floating point for reduced memory
- INT8: 8-bit integers for compact storage
- Binary: Binary vectors for maximum compression
Technical Architecture
- Built on pgrx framework for Postgres extensions
- Native Rust implementation for performance
- Integrates with PostgreSQL query planner
- WAL support for durability and replication
Getting Started
Docker (Recommended)
docker run \
--name pgvecto-rs-demo \
-e POSTGRES_PASSWORD=mysecretpassword \
-p 5432:5432 \
-d ghcr.io/tensorchord/pgvecto-rs:pg17-v0.4.0
Installation
- Pre-built binaries available
- Source compilation with cargo
- Docker images for all major PostgreSQL versions
Use Cases
- Large-scale vector search (millions to billions of vectors)
- Low-latency production applications
- Applications requiring multiple vector data types
- Hybrid search combining vectors with metadata
- PostgreSQL-native AI applications
Documentation
Comprehensive documentation at: https://docs.vectorchord.ai/
Repository
GitHub: tensorchord/pgvecto.rs Docker Hub: tensorchord/pgvecto-rs
Pricing
Free and open-source. No licensing costs for self-hosted deployments.
Loading more......
Information
Categories
Tags
Similar Products
6 result(s)Open-source PostgreSQL extension that builds on pgvector with higher-performance embedding search and cost-efficient storage. Features StreamingDiskANN index inspired by Microsoft's DiskANN algorithm. This is an OSS solution under PostgreSQL license.
PostgreSQL extension for scalable, high-performance vector search, successor to pgvecto.rs. Features RaBitQ quantization enabling 6x cost savings vs Pinecone. Fully compatible with pgvector. This is an OSS extension.
Open-source PostgreSQL extension and Python library that automates embedding generation and synchronization for RAG and semantic search applications. Features pgai Vectorizer for declarative embedding pipelines. This is an OSS solution.
Pure-Rust implementation of HNSW algorithm for approximate nearest neighbor search. Decouples graph from vector storage for flexible deployment. Supports dense floating point and quantized int8 vectors. This is an OSS library.
Open-source toolkit for developing AI applications using Postgres and pgvector. Provides managed PostgreSQL with built-in vector support, Python client (vecs), and AI features. This is a commercial managed service with OSS components.
HelixDB is a powerful, open-source graph-vector database built in Rust, designed for intelligent data storage for Retrieval-Augmented Generation (RAG) and AI applications. It combines graph database features with vector search, making it directly relevant to AI and machine learning workflows that require vector data management.