# LightRAG

> Simple and fast graph-based RAG

Retrieval-augmented generation that builds a knowledge graph from your documents for better answers.

- Source: https://github.com/HKUDS/LightRAG
- Homepage: https://arxiv.org/abs/2410.05779
- License: MIT
- Language: Python
- Stars: 39948
- Forks: 5648
- Contributors: 310
- Last commit: 2026-09-26
- Latest release: v1.5.7 (2026-09-02)
- Purpose: AI & LLM tooling, RAG & vector DBs
- Runs on: Docker / self-host, Library / SDK
- For: Personal, Enterprise

## Worth score: 86/100

How much DigGitHub recommends it, from activity, adoption, docs, license and security signals.

- Popularity: 23/25
- Momentum: 10/20
- Maintenance: 25/25
- Community: 13/15
- Readiness: 15/15

## Alternatives

- [Crawl4AI](https://diggithub.com/unclecode/crawl4ai.md): Web crawler that turns sites into LLM-ready Markdown
- [gpt-researcher](https://diggithub.com/assafelovic/gpt-researcher.md): Autonomous deep research agent for web and local documents
- [Chroma](https://diggithub.com/chroma-core/chroma.md): Open-source embedding database for AI apps
- [Firecrawl](https://diggithub.com/firecrawl/firecrawl.md): Web data API that turns sites into LLM-ready data
- [RAGFlow](https://diggithub.com/infiniflow/ragflow.md): RAG engine built on deep document understanding
- [MarkItDown](https://diggithub.com/microsoft/markitdown.md): Convert Office files and PDFs to Markdown for LLMs
- [Docling](https://diggithub.com/docling-project/docling.md): Get your documents ready for generative AI
- [OpenViking](https://diggithub.com/volcengine/OpenViking.md): Context database for AI agents

---

Source page: https://diggithub.com/HKUDS/LightRAG
Updated: 2026-10-03