graphrag

RAGFlow is an open-source retrieval-augmented generation engine that enables large models to answer questions accurately based on your private documents.

Automatically compress text before sending it to large models, saving a lot on token fees without changing answer quality.

An open-source alternative to Notebook LM. Upload materials and let AI help you summarize, answer questions, and take notes.

An open-source tool that lets AI answer questions based on your own documents and automatically organizes knowledge graphs.

Turns long documents into a tree-like outline so AI can follow the structure and reason its way to answers, without vector search.