infiniflow/infinity

An AI-native database optimized for LLM applications, offering incredibly fast hybrid search across dense vectors, sparse vectors, tensors, and full-text data.

Core Features

Incredibly fast query latency and high QPS for vector and full-text searches.

Powerful hybrid search capabilities including dense, sparse, tensor, and full-text, with advanced rerankers.

Supports a wide range of data types including strings, numerics, and vectors.

Easy to use with an intuitive Python API and a single-binary, dependency-free architecture.

GitHub Repo Documentation

Quick Start

docker run -d --name infinity -v /var/infinity/:/var/infinity --ulimit nofile=500000:500000 --network=host infiniflow/infinity:nightly

Detailed Introduction

Infinity is a cutting-edge AI-native database designed to power various LLM applications such as RAG, search, recommenders, and conversational AI. It provides robust support for rich data types like dense and sparse vectors, tensors, and full-text, enabling incredibly fast hybrid search. Its high performance, flexibility, and ease-of-use, combined with a single-binary architecture, make it an ideal solution for next-generation AI workloads.