What can I use RAGFlow for?

Upload a PDF to a Dataset and ask about it in Chat, and you get an answer with citations you can trace back. Run Knowledge Compilation on a set of documents, and it generates structured outputs such as Wikis, Mind Maps and Timelines. Ask a complex question in High thinking mode, and it breaks the question down, retrieves knowledge over several rounds and gives a grounded answer.

What do I need to use RAGFlow?

Recommended start: Docker 24.0.0 or newer, Docker Compose v2.26.1 or newer, 4 CPU cores, 16 GB RAM, 50 GB disk, plus access to an AI text model and embedding model.

RAGFlow: Ask questions about your documents and get cited answers

What it does

RAGFlow is an open-source engine for retrieval-augmented generation (RAG), which means an AI model looks things up in your documents before answering. It reads complex files such as PDFs and scanned copies, splits them into chunks using templates, retrieves relevant passages over several rounds when needed, and shows traceable citations with its answers.

How to use it

Try the hosted cloud service first, linked from the project's documentation. To run it yourself, clone the repository, start it with Docker Compose, sign in through your browser, add your models and API key, create a Dataset, upload a document, and start a Chat.

Common use cases

Upload a PDF to a Dataset and ask about it in Chat, and you get an answer with citations you can trace back.
Run Knowledge Compilation on a set of documents, and it generates structured outputs such as Wikis, Mind Maps and Timelines.
Ask a complex question in High thinking mode, and it breaks the question down, retrieves knowledge over several rounds and gives a grounded answer.

Can I use this?

Runs in Docker on your computer

Requirements It runs on your own Linux machine or server through Docker, or as the cloud service. Launching from source is meant for RAGFlow development and debugging, and it requires Go.

Caveat With the default Elasticsearch document engine, a Docker host system setting, vm.max_map_count, must be at least 262144, and a quick change resets after a reboot. Licensed under Apache-2.0.

Worth knowing Docker images cover Linux x86-64 and ARM64, and macOS is not currently supported. Resource needs grow with data volume, parsing work and local models.

Why it's here

527 people starred it on GitHub this week — 93,764 in total.

Numbers as of 8 November 2026, from BuildTube’s latest daily refresh.

Behind it

View source

BuildTube has not run or verified this project. Everything above is written from what the creator published.

Made this?

Quick questions

What is RAGFlow?: RAGFlow is an open-source engine for retrieval-augmented generation (RAG), which means an AI model looks things up in your documents before answering.
What can I use RAGFlow for?: Upload a PDF to a Dataset and ask about it in Chat, and you get an answer with citations you can trace back. Run Knowledge Compilation on a set of documents, and it generates structured outputs such as Wikis, Mind Maps and Timelines. Ask a complex question in High thinking mode, and it breaks the question down, retrieves knowledge over several rounds and gives a grounded answer.
Can I run RAGFlow on my computer?: It runs on your own computer or server after a local setup. It runs on your own Linux machine or server through Docker, or as the cloud service. Docker images cover Linux x86-64 and ARM64, and macOS is not currently supported.
What do I need to use RAGFlow?: Recommended start: Docker 24.0.0 or newer, Docker Compose v2.26.1 or newer, 4 CPU cores, 16 GB RAM, 50 GB disk, plus access to an AI text model and embedding model.
What license does RAGFlow use?: Licensed under Apache-2.0.

More from this week

DeepTutor Study your own documents with a personal AI tutor
Read documents with page-by-page citations, practice daily and get video lessons from an AI tutor.
+3,873 stars this weekAdded 7 Nov
agent-desktop Let AI agents read and click desktop apps
For developers building computer-use agents: agent-desktop lets an AI agent read and operate desktop apps on Mac, Windows or Linux from a command line.
+13,858 stars this weekAdded 3 Nov
Spec Kit Turn a written spec into plans, tasks and code
Spec Kit gives your AI coding agent a step-by-step process to build a feature, fix a bug or assess an idea.
+4,280 stars this weekAdded 2 Nov

Related projects

Claude-Mem Carry your coding agent’s project memory into the next session
Same topic: ai
+7,482 stars this weekAdd-on for an AI assistantAdded 10 Oct
starnet Watch your AI agents work inside a living pixel-art space station
Same topic: agent-harness
+148 stars this weekDesktop appAdded 25 Sept
Headroom Shrink what your AI agent reads so it uses fewer tokens
Same topic: ai
Needs a local setupAdded 21 Oct
Octop Run a private AI assistant with a team of specialized agents for your household
Same topic: agentic-ai
Needs a local setupAdded 25 Sept
harness-sdk Build a production AI agent in a few lines of Python or TypeScript
Same topic: agentic-ai
Runs in a terminalAdded 25 Sept
DeskcommCRM Run a WhatsApp sales CRM with AI agents on your own server
Same topic: ai
Runs in a terminalAdded 20 Sept

Explore by topic

Post this

For makers

Made this? Show it’s on BuildTube — add the badge to your README or embed the card on your site.

README badge (Markdown)

[![Featured on BuildTube](https://buildtube.app/badge/gh-infiniflow-ragflow.svg)](https://buildtube.app/p/gh-infiniflow-ragflow/)

Embed card (HTML)

<iframe src="https://buildtube.app/embed/gh-infiniflow-ragflow/" width="560" height="240" style="border:0;max-width:100%" loading="lazy" title="RAGFlow on BuildTube"></iframe>