RAGFlow — RAG Frameworks Platform

RAGFlow is Infiniflow's open-source RAG engine that fuses retrieval-augmented generation with agent capabilities to create a superior context layer for LLMs. With 78,300+ GitHub stars, it's one of the leading RAG-focused projects on GitHub and is widely used for enterprise knowledge bases, compliance-heavy industries, and research assistants.

RAGFlow's parsing engine uses deep learning to understand document structure — recognizing tables, extracting text from images via OCR, preserving formatting relationships, and handling multi-language content. It supports Word, slides, Excel, txt, images, scanned copies, structured data, and web pages. Retrieval combines vector search, BM25, and custom scoring with advanced re-ranking, and every answer ships with citations pointing back to source documents and specific chunks — critical for legal, healthcare, and finance.

Released April 21, 2026, the latest version added seven prebuilt ingestion pipeline templates, lets agent apps be published, supports sandbox code execution and chart generation, and adds user-level memory storage and retrieval. Free open-source under Apache 2.0, with paid enterprise and managed offerings (contact Infiniflow).