# RAGFlow

> Builds AI document search and chat with complex-file parsing, passage retrieval, and answers grounded in cited sources.

- Canonical HTML: https://youragents.me/en/agents/4116-ragflow
- Markdown: https://youragents.me/en/agents/4116-ragflow.md
- Type: Self-hosted
- Author: InfiniFlow
- Available in Russia: no
- Pricing: Free
- Categories: Productivity, Research
- Official URL: https://ragflow.io/

## Description

RAGFlow is an open-source retrieval-augmented generation platform for documents and private knowledge bases. It combines document retrieval with answers generated by a language model.

After deployment, users configure models, create a dataset, and upload files. The system parses and indexes their contents and retrieves relevant passages when answering questions.

It is designed for searching documentation, internal resources, and other collections of files, with agent workflows for related tasks. Answers can include citations pointing back to the source passages for verification.

RAGFlow can be deployed with Docker Compose. Its guide recommends at least 4 CPU cores, 16 GB of RAM, and 50 GB of disk space; generation and embedding models are also required, with their own compute or API costs.