ragflow

Project Overview

RAGFlow is a leading open-source Retrieval-Augmented Generation (RAG) engine that combines cutting-edge RAG technology with Agent capabilities, providing a superior context layer for LLMs. It offers a streamlined RAG workflow adaptable to enterprises of any scale.

Key Features

• "Quality in, quality out": Deep document understanding-based knowledge extraction from unstructured data
• Agent workflows: Supports agentic workflow and MCP (Model Context Protocol)
• Multi-source sync: Supports Confluence, S3, Notion, Discord, Google Drive
• Multimodal support: Supports multimodal models for understanding images within PDF or DOCX files
• Multiple parsing methods: Supports MinerU and Docling
• Multi-channel integration: Supports Feishu, Discord, Telegram, Line, and more
• Agent memory: Supports "Memory" for AI agents

Target Audience

• Enterprises and developers building RAG applications
• Data scientists who want to combine LLMs with private data
• Teams interested in agentic workflows

Quick Start

1. Try the cloud service: https://cloud.ragflow.io
2. Check the official documentation
3. See the roadmap

Core Advantages

• Deep document understanding for complex data
• Deep integration of RAG and Agent capabilities
• Supports multiple data sources and channel integrations

System Requirements

• OS: Cross-platform (Docker deployment supported)