Open source
LightRAG: Graph Answers From Your Docs
LightRAG is an open source graph based retrieval augmented generation project from HKUDS. It reads your documents, builds a knowledge graph of entities and relations, and answers questions using that graph plus vector search. Learn more at
Watch on YouTube · Watch on TikTok
The story
What the video said
This open source GitHub project turns a pile of documents into a knowledge graph you can question. It reads your files, finds the people and places inside, and links them together. You already loaded your notes into a chat tool and asked it something. The answer missed a link between two names. You want to ask again and get the full picture this time. LightRAG is an open source library and server for this. It stores your text as a graph and as vectors, then searches both when you ask a question. The project page shows a graph viewer in its web app. You can see your documents drawn as linked nodes, and you can pick a search mode like local or global. It is free and open source, so you can run it on your own machine. Would you use this workflow to search your own documents?
This page was generated from the Ming Dao AI publication record.
Ming Dao AI is independent and unaffiliated with GitHub or the featured project. Read our editorial policy.