RAGFlow review

Q: How do I install RAGFlow?

Run git clone https://github.com/infiniflow/ragflow.git. The install section above has the other options from the project's README.

RAGFlow is an open-source Retrieval-Augmented Generation engine with agent capabilities that turns complex documents and data sources into context for LLMs.

Updated Nov 1, 2026. Review written Oct 25, 2026 from the project's own sources; usage data refreshed every Monday.

TL;DR

RAGFlow is an open-source RAG engine that combines document parsing, retrieval and agent workflows. It is for developers building question-answering or agent systems over their own documents and data. It stands out for deep document understanding, template-based chunking, Knowledge Compilation and traceable citations, and it runs self-hosted with Docker or as a cloud service.

93.3kGitHub stars
+497new stars, week of Oct 25
#16in RAG, data and memory this week

RAGFlow at a glance

Category: RAG, data and memory
Maintained by: InfiniFlow (organization, @infiniflow on GitHub)
License: Apache 2.0
Language: Go, TypeScript, C++
Repository: github.com/infiniflow/ragflow
Website: ragflow.io
Latest release: v0.27.3 on GitHub, Oct 23, 2026
First published: Dec 12, 2023

What RAGFlow does

RAGFlow extracts knowledge from Word, PowerPoint, Excel, TXT, images, PDFs, scanned copies, structured data and web pages.
RAGFlow offers template-based chunking with visualization of chunks, so people can review and adjust them.
RAGFlow returns answers with key references and traceable citations to support grounded responses.
RAGFlow performs multi-step agentic retrieval that breaks down complex questions, retrieves knowledge and verifies evidence.
RAGFlow's Knowledge Compilation generates Wikis, Graphs, Trees, PageIndex, Mind Maps, Timelines and Skills from documents and datasets.
RAGFlow combines vector search, BM25 full-text search and custom scoring with fused re-ranking.
RAGFlow builds agents in visual workflows that combine RAG, tools, web search and MCPs.

How it works

RAGFlow runs as a set of services: Admin, Ingestor, Syncer and API. The API serves the Web UI and SDKs. Docker Compose starts these services with dependencies such as MySQL, MinIO, NATS, Kvrocks and Elasticsearch or Infinity. You add an LLM, an embedding model and a reranker with their service addresses and API keys. Version 1.0 is a rewrite in Go.

How to install and start using RAGFlow

Clone the repository

git clone https://github.com/infiniflow/ragflow.git

Check out the release tag

git checkout v1.0.0-rc2

Start with Docker Compose (from ragflow/docker)

docker compose -f docker-compose.yml up -d

Requirements

Docker ≥ 24.0.0 and Docker Compose ≥ v2.26.1.
Recommended starting configuration: 4 CPU cores, 16 GB RAM and 50 GB free disk space.
API keys and service addresses for an LLM, an embedding model and a reranker.
Building from source requires Go 1.27, Clang 20, LLD 20, CMake ≥ 4.0 and PCRE2 development files.

First steps

If you use Elasticsearch, set vm.max_map_count on the Docker host to at least 262144.
After startup, check readiness with curl -f http://localhost/api/v1/system/healthz and wait for an HTTP 200 response.
Open http://IP_OF_YOUR_MACHINE in a browser and log in to the Web UI.
On the model provider page, add an LLM, an embedding model and a reranker with model name, service address and API key.

Who RAGFlow is for

Teams building question answering over complex documents such as scanned PDFs, spreadsheets and presentations.
Developers who need answers with traceable citations to reduce hallucinations.
Organizations that want a self-hosted RAG and agent platform with configurable models.
Domain workflows such as investment research, legal precedent analysis and maintenance support.

Limitations to know

RAGFlow's Docker platform support does not currently include macOS.
In the open-source 1.0 release, DeepDoc uses CPU inference for layout analysis, OCR and table recognition.
RAGFlow 1.0.0 is at release-candidate stage, so the current recommended tag is a preview release.
RAGFlow needs a recommended 16 GB RAM, and local models or optional components may need more.

Is RAGFlow growing? Usage data

In the week of Oct 25–31, 2026, RAGFlow ranked #140 on the Devtoolwatch Board and #16 in RAG, data and memory, with a momentum score of 55 out of 100.

New GitHub stars per week for the last 12 weeks, from GitHub's stargazer history (today's stargazers by the week they starred).

Show the weekly numbers

New GitHub stars per week for RAGFlow
Week of	New stars
Aug 9, 2026	1.5k
Aug 16, 2026	534
Aug 23, 2026	609
Aug 30, 2026	537
Sep 6, 2026	534
Sep 13, 2026	466
Sep 20, 2026	355
Sep 27, 2026	339
Oct 4, 2026	395
Oct 11, 2026	463
Oct 18, 2026	469
Oct 25, 2026	497

On Hacker News

RAGFlow is an open-source RAG engine based on OCR and document parsing, 230 points and 53 comments (Apr 1, 2024)
RAGFlow: Modern Agentic RAG Based on Graph, 13 points and 5 comments (Jul 11, 2024)
DeepSeek-V2 integrated, RAGFlow v0.5.0 is released, 7 points and 0 comments (May 8, 2024)

The Board ranks tools by four signals each week: new stars, new stars for the repo's size, added downloads and download growth beyond the market. How the ranking works.

RAGFlow alternatives

Other RAG, data and memory tools on the Board in the week of Oct 25–31, 2026, by momentum:

RAG, data and memory ranked by momentum, the week of Oct 25–31, 2026
Rank	Tool	Stars	Downloads/wk
#1	XERJXERJ is a local-first search engine and MCP server for AI agents, packaged as one Rust binary with an Elasticsearch-compatible API.	17.9k	–
#2	MemvidMemvid is a single-file memory layer for AI agents that packages data, embeddings, search indexes, and metadata into one portable .mv2 file.	17.3k	–
#3	LEANNLEANN is a Python vector database and RAG library that recomputes embeddings on demand to keep indexes small on personal devices.	13.7k	–
#4	GraphifyGraphify is an AI coding assistant skill that turns a codebase, with its docs, PDFs, images and videos, into a queryable knowledge graph.	140k	480.7k
#5	MarkItDownMarkItDown is a lightweight Python utility from Microsoft that converts PDFs, Office documents, images, audio and other files to…	193.2k	4.7M
#6	FirecrawlFirecrawl is an API and open source project to search, scrape, crawl, and interact with web pages, returning LLM-ready markdown or JSON…	197.4k	2.9M

All RAG, data and memory tools, ranked

Recent RAGFlow releases

nightly (pre-release), Dec 1, 2025
v0.27.3, Oct 23, 2026
v1.0.0-rc2, Oct 16, 2026
v1.0.0-rc1, Sep 29, 2026
v0.27.2, Sep 10, 2026

RAGFlow FAQ

Is RAGFlow free and open source?

Yes. RAGFlow's source code is on GitHub (infiniflow/ragflow) under the Apache 2.0 license, so you can use it for free. Its website also links to a pricing page (https://ragflow.io/#pricing-plan) for any paid plans.

How do I install RAGFlow?

Run git clone https://github.com/infiniflow/ragflow.git. The install section above has the other options from the project's README.

Is RAGFlow still maintained?

Yes. The infiniflow/ragflow repository was last updated on Oct 25, 2026, and the latest release, v0.27.3 on GitHub, came out on Oct 23, 2026.

How popular is RAGFlow?

RAGFlow has 93.3k GitHub stars as of Nov 1, 2026. It gained 497 new stars in the week of Oct 25–31, 2026, against a 12-week average of 559 a week. That made it #16 among RAG, data and memory tools on the Devtoolwatch Board.

RAGFlow vs LlamaIndex: how do they compare?

RAGFlow and LlamaIndex are both tracked by Devtoolwatch. RAGFlow has 93.3k GitHub stars and LlamaIndex has 52.8k. LlamaIndex review.

RAGFlow vs LightRAG: how do they compare?

RAGFlow and LightRAG are both tracked by Devtoolwatch as RAG, data and memory tools. RAGFlow has 93.3k GitHub stars and LightRAG has 40.5k. LightRAG is an open-source Python framework for retrieval-augmented generation that combines knowledge graphs with vector embeddings for indexing and retrieval. LightRAG review.

What are the best alternatives to RAGFlow?

Other RAG, data and memory tools on the Devtoolwatch Board in the week of Oct 25–31, 2026, by momentum: XERJ (#1), Memvid (#2), LEANN (#3), Graphify (#4), MarkItDown (#5). The alternatives table above has their numbers.

Is RAGFlow open source?

Yes. RAGFlow is an open-source RAG engine released under the Apache-2.0 license. You can self-host it with Docker or build it from source. InfiniFlow also offers a hosted cloud service at cloud.ragflow.io.

What is RAGFlow used for?

RAGFlow is used to build RAG and agent systems over your own data. It parses complex documents, chunks and indexes them, retrieves relevant context and generates answers with citations. The homepage shows example workflows for investment research, legal precedent analysis and manufacturing maintenance.

How does RAGFlow work?

RAGFlow ingests and parses documents, chunks them with templates, and indexes them for hybrid vector and full-text search with re-ranking. At query time, agentic retrieval can break down questions and verify evidence before an LLM generates a cited answer. Admin, Ingestor, Syncer and API services run the system.

Which LLMs does RAGFlow work with?

RAGFlow works with the LLMs, embedding models and rerankers you configure. You add each one on the model provider page with its name, service address and API key. Recent releases added providers such as DemonRoute, Opper and Synthorai, plus AWS Bedrock API key authentication.

Which data sources can RAGFlow ingest?

RAGFlow ingests Word, PowerPoint, Excel, TXT, images, PDFs, scanned copies, structured data and web pages. It also ingests websites through sitemaps. Connectors include Google BigQuery with incremental sync and Azure DevOps.

Does RAGFlow support MCP?

Yes. RAGFlow agent workflows can integrate MCPs along with RAG, tools and web search in its visual workflow builder.

Sources

RAGFlow README badge

Maintainers can show the tool's weekly Board rank in their README:

[![Devtoolwatch](https://index.devtoolwatch.com/badge/ragflow.svg)](https://index.devtoolwatch.com/tools/ragflow/)

How this page is made: Devtoolwatch writes each review from the project's README, releases, package metadata and website, with an AI model, then checks every claim against the source it cites and drops what the sources don't support. The numbers come from Devtoolwatch's daily GitHub snapshots, GitHub's stargazer history, and the npm and PyPI download APIs. Spotted an error? Email [email protected].