Overview
FastGPT is an AI agent building platform that gives you out-of-the-box data processing and model calling. Instead of wiring everything yourself, you import documents into a knowledge base and connect logic with a visual Flow editor to build complex question-answering apps.
It runs as a self-hosted service you start with Docker, and there is also a hosted cloud version at fastgpt.io. Once running, you reach the web app on localhost:3000 and sign in with the default root account to start building.
Because it exposes an OpenAPI interface aligned with the GPT chat format, apps you build in FastGPT can be embedded in a page, shared through a login-free window, or called from your own backend like any other chat API.
What it does
- Visual Flow editor for orchestrating dialogue and plugin workflows, including basic RPA nodes and a planning agent mode
- Built-in knowledge base supporting txt, md, html, pdf, docx, pptx, csv, and xlsx files, plus URL reading and CSV batch import
- Hybrid retrieval and reranking across multiple knowledge bases that can be reused and mixed together
- OpenAPI interface with a completions endpoint aligned to the GPT chat format, plus CRUD APIs for knowledge bases and conversations
- Bidirectional MCP support and visual model configuration for connecting different models
- Operations tools including login-free share windows, one-click iframe embedding, dialogue record review, and a template marketplace
Getting started
FastGPT is deployed with Docker. The install script pulls the configuration files, then Docker Compose brings the service up.
Pull the configuration files
Run the install script in your terminal and follow the prompts to fetch the configuration.
bash <(curl -fsSL https://doc.fastgpt.cn/deploy/install.sh)Start the service
Bring the stack up in the background with Docker Compose.
docker compose up -dOpen and sign in
Once everything is running, open http://localhost:3000 in your browser. The default account is root and the password is 1234.
Commands and code are distilled from the project's own documentation — always check the official repo for the latest.
When to use it
- Build a document question-answering assistant by importing PDFs, Word files, or web pages into a knowledge base and chatting over them with hybrid retrieval
- Design multi-step conversational workflows visually with the Flow editor, combining knowledge lookups, plugins, and user interaction nodes
- Embed a finished agent into an existing website through an iframe or a login-free share link, or call it from your backend via the OpenAPI chat endpoint
How FastGPT compares
FastGPT alongside other open-source rag frameworks & platforms tools AI/TLDR tracks, ranked by GitHub stars.
| Tool | Stars | What it does |
|---|---|---|
| Dify | ★ 146k | An open-source platform with a visual workflow builder for creating LLM and RAG applications without writing much code. |
| RAGFlow | ★ 83.2k | A RAG engine built around deep document understanding that turns complex files into a grounded, citation-backed question-answering layer. |
| Context7 | ★ 57.7k | Context7 pulls current, version-specific documentation and code examples for any library and feeds them into your LLM, available as a CLI skill or an MCP server. |
| Quivr | ★ 39.2k | Quivr is an open-source RAG framework that ingests your documents and answers questions about them, working with any LLM and any file type. |
| LightRAG | ★ 36.8k | A graph-based RAG system that builds an entity-and-relationship knowledge graph for fast retrieval and easy incremental updates. |
| GraphRAG | ★ 33.9k | Microsoft's graph-based RAG system that extracts a knowledge graph from documents to answer broad, multi-document questions. |
| PageIndex | ★ 33.2k | PageIndex turns long PDFs into a table-of-contents tree and uses LLM reasoning to retrieve relevant sections, with no vector database and no chunking. |
| FastGPT | ★ 28.6k | Build knowledge-based AI agents with a visual workflow editor |