Skip to content
  1. Home
  2. AI tools
  3. LightRAG

LightRAG

Graph-based RAG library and server with pluggable storage backends

What it is

A research-originated RAG system that extracts entities and relations into a knowledge graph and combines it with vector retrieval. It ships a Python library, an API server with web UI and several chunking strategies, and recently absorbed multimodal parsing through MinerU or Docling services. For developers exploring graph-enhanced retrieval.

Good for

  • Build a knowledge-graph index from documents
  • Run a RAG API server with a web UI
  • Compare chunking strategies on your corpus

Install or run

uv tool install "lightrag-hku[api]"

Command as given in the project README. Read the repository before you run anything on your machine.

What you need

Python with uv; LLM and embedding model access

Worth knowingIndexing calls an LLM to extract entities, so large corpora cost time and tokens; the project changes quickly.

Sources

The summary is ours, written from the README. Spotted an error? See how we work.

More for data & research

See the topic →
  • MarkItDown

    microsoft/markitdown

    Lightweight Python utility that converts files to Markdown for LLMs

    ToolGitHub stars: 195k
  • RAGFlow

    infiniflow/ragflow

    Self-hosted RAG engine with document parsing and agent templates

    ToolGitHub stars: 94k
  • Docling

    docling-project/docling

    Parse PDFs, Office files and more into structured data for LLMs

    ToolGitHub stars: 71k
  • serviette

    pathwaycom/serviette

    No-code RAG server that keeps answers in sync with a document folder

    ToolGitHub stars: 59k