infiniflowAI apps
ragflow
An open-source RAG engine fusing deep document understanding and agent capabilities to build grounded context layers for LLMs.
- Backend
- Full stack
- DevOps
- Data
- Automation
- Data analysis
- Productivity
- Web
- API/SDK
- Linux
- macOS
- Windows
- Browser
- Self-hostable
- Runs locally
- Docker supported
- Needs API key
- Popularity
- 91.6k Stars
- GitHub stars
- Recent activity
- 9/30/2026
- Updated in the last 30 days
- License
- APACHE-2.0
- Permissive
Why it matters
We look beyond stars: what problem it solves, whether it creates real utility, and what makes its approach worth noticing.
Problem
For enterprise knowledge and complex documents, the bottleneck is rarely just having a vector store; document understanding, chunking, retrieval, and context construction determine whether RAG reduces or amplifies hallucinations and omissions.
Practical value
RAGFlow combines deep document understanding, retrieval-augmented generation, and agent capabilities in one system so teams can build grounded, traceable LLM context layers from raw documents faster.
Innovation / differentiation
Rather than wrapping a vector-search endpoint, it puts complex-document understanding and RAG quality at the center and connects that layer upward into agent workflows, covering the full path from knowledge to model context.
Leverage potential
A reliable context layer can be reused by Q&A, search, support, research assistants, and other agents, so its leverage extends well beyond a single chat interface.
Why now
As base models improve, enterprise AI bottlenecks are shifting toward reliably injecting private knowledge into context, while agents increase demand for traceable knowledge and long-document understanding.
Community activity
In the last 90 days there were 447 new issues and 3444 pull requests; the bounded issue/PR samples include 60 issue authors and 26 PR contributors, with at least 5 releases.
Maintainer responsiveness
The 96-issue window sample had a 46% close rate, and the 80-pull-request sample had a 85% merge rate. Maintainer-comment observations covered only 6% of that issue sample, so response rate and first-response speed are not reported.
Key highlights
- Knowledge extraction from unstructured data via deep document understanding
- Template-based intelligent text chunking with visualization support
- Grounded citations and traceability to reduce hallucinations
Quick start
How it is installed, how hard it is, and where to start.
Where it runs
Docker
Difficulty
Medium — some setup needed
- 01Check and set the vm.maxmapcount host parameter to at least 262144.
- 02Clone the RAGFlow repository to your local machine.
- 03Navigate to the docker directory and start services using docker compose.
- 04Check container logs to ensure initialization is complete, then access the server IP in your web browser.
Best for
- Teams that want data on their own servers
- Developers who want to try it on their machine
- People who prefer Docker deploys
More about it
RAGFlow is an open-source Retrieval-Augmented Generation (RAG) engine that fuses RAG with agent capabilities to build a superior context layer for LLMs. It focuses on extracting knowledge from complex document layouts through deep document understanding, template-based chunking, and multi-path retrieval fusion with traceable citations.
The system provides streamlined orchestration and self-hosting support via Docker, allowing developers to connect custom LLM APIs and establish production-ready AI applications.
Sources
Each field shows its status and source — expand to review.
12 · Expand
Sources
Each field shows its status and source — expand to review.
capability tags
Verifiedautomation, data_analysis, productivity
Source: admin_cms · cms editor · 8/17/2026
Latest release
Verifiedv1.0.0-rc1
Source: GitHub API · latest_release=v1.0.0-rc1 · 10/1/2026
License
VerifiedApache-2.0
Source: GitHub API · license.spdx_id=Apache-2.0 · 10/1/2026
needs api key
VerifiedYes
Source: admin_cms · cms editor · 8/15/2026
One-liner
Verified{"en":"An open-source RAG engine fusing deep document understanding and agent capabilities to build grounded context layers for LLMs.","zh":"基于深度文档解析与智能体能力的开源 RAG 引擎，用于构建具备可追溯引用的大模型上下文层。"}
Source: admin_cms · cms editor · 8/15/2026
Platforms
Verifiedlinux, macos, windows, browser
Source: admin_cms · cms editor · 8/15/2026
Category hint
Inferred from materialsai-apps
Source: Project README · hint=ai-apps · 8/15/2026
product forms
Verifiedweb, api_sdk
Source: admin_cms · cms editor · 8/17/2026
role tags
Verifiedbackend, fullstack, devops, data
Source: admin_cms · cms editor · 8/17/2026
supports docker
VerifiedYes
Source: Repository file · dockerfile=true; compose=true · 10/1/2026
supports local
VerifiedYes
Source: admin_cms · cms editor · 8/15/2026
supports self host
VerifiedYes
Source: admin_cms · cms editor · 8/15/2026
Related projects
Other verified projects matched by category, capabilities, and intended roles.
ouroboros
Ouroboros is an open-source, general-purpose AI agent with durable cross-restart memory that coordinates specialist swarms and rewrites its own implementation.
needle
A small local-model Python package for tool calling and structured extraction, with offline inference, confidence thresholds, LoRA fine-tuning, and model export.
sutando
An open-source local AI agent that responds via voice and screen by day, and runs an autonomous build loop by night.
LobsterAI
A desktop agent for local files, terminal, browser, and office workflows with multi-agent setup, skills, MCP, scheduled tasks, and IM remote control.
hermes-agent
A personal AI agent that runs locally or on remote infrastructure with terminal, messaging, memory, skills, and scheduled automation.
Box
An advanced, fully offline on-device AI suite for Android supporting chat, image generation, and voice features.