Ragflow

RAGFlow is the open-source Retrieval-Augmented Generation (RAG) engine built for enterprise, delivering reliable context via deep document understanding and high-precision hybrid search.

RAGFlow functions as the leading open-source RAG engine, focusing on superior context delivery for AI agents. It leverages deep document understanding and advanced ETL (Extract, Transform, Load) pipelines to process multi-format data (PDFs, images, etc.), structuring it for optimal retrieval. The system employs high-precision hybrid search, combining vector, BM25, and custom scoring for unmatched accuracy. Crucially, it minimizes LLM 'hallucinations' by providing grounded, citable answers, and offers a unified platform for orchestrating RAG, tools, and agents within visual workflows (e.g., for equity research or legal analysis).

https://ragflow.io