RAGFlow is an open-source retrieval-augmented generation (RAG) engine that turns messy documents into a searchable knowledge base for chat and agents.
In plain words
RAG systems work only as well as the text they retrieve. RAGFlow focuses on reading hard documents properly, using layout analysis, OCR and table recognition on Word files, slides, spreadsheets, scans and web pages. It splits them into chunks with templates you can inspect and adjust, and answers come with traceable citations. You run it yourself with Docker, connecting your own LLM and embedding model, or use its cloud service.
A quick example
Imagine a company loading scanned contracts and spreadsheets into RAGFlow. Staff then ask a chat assistant which contracts include a penalty clause, and each answer links to the passages it used.
Where you'll meet it
- Self-hosted question answering over company documents.
- Building a context layer for AI agents.
Related
Sources
- DocsRAGFlow README, InfiniFlow (GitHub) · read 12 Oct 2026
- DocsRAGFlow quick start, RAGFlow docs · read 29 Sept 2026
Checked by WhatIsAI against the sources above · 12 Oct 2026. Spotted an error? Report it.