What do you want to do?
3 use cases
-
Run a knowledge-graph RAG server with a web UI
Install the LightRAG Server, point it at your models and query documents through the WebUI.
- REPO
HKUDS/LightRAG - Claude Code
- Codex
- Cursor
- Research
- Intermediate
View prompt
Install HKUDS/LightRAG Server with uv, configure .env for <LLM> and <embedding model>, bind it to 127.0.0.1, and load <documents>. Then ask <question> and show me the entities and relations it used.
- REPO
-
Deploy LightRAG with Docker Compose
Clone the repo, fill in .env and bring the whole stack up with one command.
- REPO
HKUDS/LightRAG - Claude Code
- Codex
- Cursor
- Research
- Intermediate
View prompt
Deploy HKUDS/LightRAG with Docker Compose on <server>, configure .env for <LLM> and <embedding model> with authentication turned on, and confirm the WebUI loads. Report the URL and any startup warnings.
- REPO
-
Configure LightRAG safely with the setup wizard
Use the interactive make targets to generate .env, add storage backends and audit security.
- REPO
HKUDS/LightRAG - Claude Code
- Codex
- Cursor
- Research
- Intermediate
View prompt
Set up HKUDS/LightRAG from source using the setup wizard: run make env-base for <LLM>, make env-storage for <backend>, make env-server with auth on, then make env-security-check. Fix anything the check flags before starting lightrag-server.
- REPO
Quick start with LightRAG
- Check you have what it needs: Python with uv (recommended) or pip, or Docker Compose; An LLM and an embedding model configured in .env; Optional: storage backends such as PostgreSQL, Neo4j or OpenSearch.
- Open HKUDS/LightRAG on GitHub and read the README and license.
- Install the server as a tool with
uv tool install "lightrag-hku[api]" - Copy the example config with
cp env.example .envand set your LLM and embedding settings - Start it with
lightrag-server - Open the WebUI to insert documents, query them and visualize the graph
These steps come from our top use case, Run a knowledge-graph RAG server with a web UI, which also has a copy-paste agent prompt.
Use this in your AI agent
Ask your agent
Using the Deduped MCP, get the setup steps for HKUDS/LightRAG
Add Deduped to Claude Code once
claude mcp add --transport http deduped https://dedupedai.com/mcp
Every Monday: the 5 hottest repos and exactly how to use them. Free, one email a week, unsubscribe anytime.
Get them freeAbout this repo
- Who it is forBuilders who want RAG that understands relationships between entities across documents, with a ready-made server and UI.
- Why it is hot40k stars, talked about in 2 AI sources we track
- LanguagePython
- Sourcegithub.com/HKUDS/LightRAG ยท Back to Hot repos
Requirements
- Python with uv (recommended) or pip, or Docker Compose
- An LLM and an embedding model configured in .env
- Optional: storage backends such as PostgreSQL, Neo4j or OpenSearch
Add this badge to your README
Maintain LightRAG? Show readers where to find these use cases.
Markdown
[](https://dedupedai.com/use-cases/hkuds-lightrag/?utm_source=github&utm_medium=badge&utm_campaign=HKUDS-LightRAG)
HTML
<a href="https://dedupedai.com/use-cases/hkuds-lightrag/?utm_source=github&utm_medium=badge&utm_campaign=HKUDS-LightRAG"><img src="https://dedupedai.com/badge/HKUDS/LightRAG.svg" alt="Use cases on Deduped" height="20"></a>
A static image, no tracking script. About the badge
Questions about LightRAG
What is LightRAG?
LightRAG is a lightweight knowledge-graph RAG framework from HKU, positioned as an efficient alternative to Microsoft GraphRAG. It builds both a knowledge graph and vector embeddings from your documents, ships a server with a WebUI to insert, query and visualize knowledge, supports rerankers, citations, document deletion with graph regeneration, multimodal parsing through MinerU or Docling, and storage backends such as PostgreSQL, Neo4j, MongoDB and OpenSearch. It is an open-source project on GitHub: https://github.com/HKUDS/LightRAG.
How do I use LightRAG?
Install the server as a tool with uv tool install "lightrag-hku[api]"; then copy the example config with cp env.example .env and set your LLM and embedding settings; start it with lightrag-server. The use cases on this page give the full steps and a copy-paste agent prompt for each job.
Who is LightRAG for?
Builders who want RAG that understands relationships between entities across documents, with a ready-made server and UI.
What can I do with LightRAG?
3 use cases so far, for example: Run a knowledge-graph RAG server with a web UI; Deploy LightRAG with Docker Compose; Configure LightRAG safely with the setup wizard. Each has steps and a copy-paste agent prompt.
What do I need to run LightRAG?
Python with uv (recommended) or pip, or Docker Compose; An LLM and an embedding model configured in .env; Optional: storage backends such as PostgreSQL, Neo4j or OpenSearch.
What are alternatives to LightRAG?
Other hot AI repos in Research: Deep Tutor, Local Deep Research, Project Nomad, Nanochat, We Knora.
Is LightRAG safe to run?
Deduped does not own, audit or vouch for it. Read the README and license, run it on your own systems first and review every command before you run it.