RAGFlow

RAGFlow is an open-source retrieval-augmented generation (RAG) engine that combines deep document understanding with agent capabilities, so AI apps can answer from your documents. You can run it yourself or use its cloud service.

RAGFlow extracts knowledge from complex, unstructured documents, uses explainable template-based chunking, and gives grounded answers with citations. Its Knowledge Compilation turns documents and datasets into wikis, graphs, mind maps, timelines and skills, and Agentic RAG adds multiple thinking modes. It can also ingest websites through sitemaps.

RAGFlow is one of the strongest open-source options when answer quality depends on parsing messy documents well. It is Apache-licensed.

Stars

0

Forks

0

Commits

0

Repository

infiniflow/ragflow

• VectaraHosted RAG and search platform.
• Azure AI SearchMicrosoft's search and retrieval service for AI apps.
• DifyNo. 021Open-source platform for building LLM apps.