FastGPT
FastGPT is a knowledge-based platform built on the LLMs, offers a comprehensive suite of out-of-the-box capabilities such as data processing, RAG retrieval, and visual AI workflow orchestration, letting you easily develop and deploy complex question-
What is FastGPT?
FastGPT is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to fastgpt is a knowledge-based platform built on the llms, offers a comprehensive suite of out-of-the-box capabilities such as data processing, rag retrieval, and visual ai workflow orchestration, letti...
FastGPT is a knowledge-based platform built on the LLMs, offers a comprehensive suite of out-of-the-box capabilities such as data processing, RAG retrieval, and visual AI workflow orchestration, letting you easily develop and deploy complex question-
This server falls under the Business Applications category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- FastGPT is a knowledge-based platform built on the LLMs, off
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx fastgptConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use FastGPT
FastGPT is a self-hosted, open-source knowledge base and AI agent platform built on top of LLMs, offering visual workflow orchestration, hybrid RAG retrieval with reranking, multi-format document ingestion (PDF, DOCX, Excel, Markdown, and more), and a full OpenAPI interface. It supports bidirectional MCP integration — FastGPT can both act as an MCP server exposing its knowledge bases as tools, and consume external MCP servers inside its visual agent workflows. Teams use it to build internal question-answering systems, automated data processing pipelines, and customer-facing AI assistants backed by proprietary knowledge.
Prerequisites
- Docker and Docker Compose installed on your host machine
- At least 2 GB of free RAM and 10 GB of disk space for the full stack
- An API key for an LLM provider (e.g. OPENAI_API_KEY for OpenAI, or a compatible endpoint)
- MongoDB and PostgreSQL with pgvector extension (included in the Docker Compose stack)
- An MCP-compatible client if you want to connect to FastGPT's MCP server from an external tool
Deploy FastGPT using the automated install script
Run the official one-liner to download the Docker Compose stack and start all services. This is the fastest way to get a working instance.
bash <(curl -fsSL https://doc.fastgpt.cn/deploy/install.sh)
docker compose up -dAccess the FastGPT web interface
Open the web UI in your browser. The default credentials are root/1234 — change them immediately after first login.
# Open browser at:
http://localhost:3000
# Default login: root / 1234Configure your LLM provider
In the FastGPT admin panel, go to Settings > Model Provider and add your LLM API key and endpoint. FastGPT supports OpenAI-compatible APIs, so you can point it at any compatible model server.
Create a knowledge base and import documents
In the FastGPT interface, create a new Knowledge Base, upload your documents (PDF, DOCX, txt, csv, etc.), and run the indexing process to make them searchable.
Build an application with visual workflow orchestration
Use FastGPT's visual Flow editor to create an agent that retrieves from your knowledge base, calls external APIs, or invokes MCP tools. Connect nodes for intent recognition, knowledge retrieval, LLM response generation, and output formatting.
Access the FastGPT OpenAPI or MCP endpoint
FastGPT exposes an OpenAPI-compatible chat endpoint and, when MCP mode is enabled, an MCP server endpoint that lets external AI clients query your knowledge bases as tools.
# FastGPT OpenAPI chat endpoint (compatible with OpenAI SDK)
curl http://localhost:3000/api/v1/chat/completions \
-H 'Authorization: Bearer YOUR_FASTGPT_API_KEY' \
-H 'Content-Type: application/json' \
-d '{"model": "YOUR_APP_ID", "messages": [{"role": "user", "content": "Hello"}]}'FastGPT Examples
Client configuration
MCP client configuration for connecting to a self-hosted FastGPT MCP server endpoint.
{
"mcpServers": {
"fastgpt": {
"command": "npx",
"args": ["fastgpt"],
"env": {
"FASTGPT_BASE_URL": "http://localhost:3000",
"FASTGPT_API_KEY": "your_fastgpt_api_key"
}
}
}
}Prompts to try
Prompts for building and querying knowledge-based applications through FastGPT's AI workflow platform.
- "Search my company knowledge base for the refund policy and provide a detailed answer for a customer who wants to return a product after 45 days"
- "Process the uploaded CSV of customer feedback and categorize each row by sentiment and topic"
- "Build a workflow that answers technical support questions by first searching the docs knowledge base, then escalating to a human if confidence is low"
- "Summarize all documents uploaded to the 'Q3 Reports' knowledge base and identify the three most important business insights"
- "Create a QA dataset from the product manual by generating question-answer pairs for each section"Troubleshooting FastGPT
Docker Compose starts but FastGPT shows a database connection error
MongoDB and PostgreSQL containers may still be initializing. Wait 60 seconds and refresh. Check container health with 'docker compose ps' and view logs with 'docker compose logs mongo' or 'docker compose logs pg'. Ensure no existing services are occupying ports 27017 or 5432.
Knowledge base indexing fails or documents don't appear in search results
Indexing requires a working embedding model endpoint. Confirm your LLM provider API key is set and the embedding model specified in FastGPT's model settings is available and not rate-limited. Check the FastGPT server logs for embedding API errors.
Workflow nodes produce unexpected outputs or agent loops indefinitely
Enable the call-chain debug view in the FastGPT flow editor to trace execution step by step. Add explicit output formatting nodes or add a maximum-iteration limit to loop nodes. Check the LLM model settings — some models require specific system prompt formatting for tool-use to work correctly.
Frequently Asked Questions about FastGPT
What is FastGPT?
FastGPT is a Model Context Protocol (MCP) server that fastgpt is a knowledge-based platform built on the llms, offers a comprehensive suite of out-of-the-box capabilities such as data processing, rag retrieval, and visual ai workflow orchestration, letting you easily develop and deploy complex question- It connects AI assistants to external tools and data sources through a standardized interface.
How do I install FastGPT?
Follow the installation instructions on the FastGPT GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with FastGPT?
FastGPT works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is FastGPT free to use?
Yes, FastGPT is open source and available under the NOASSERTION license. You can use it freely in both personal and commercial projects.
FastGPT Alternatives — Similar Business Applications Servers
Looking for alternatives to FastGPT? Here are other popular business applications servers you can use with Claude, Cursor, and VS Code.
n8n
★ 189.1kA comprehensive MCP server that provides full control over n8n automation workflows through natural language. It offers 43 tools for managing workflows, executions, credentials, and data tables, with safety features like write-mode protection and dou
LobeHub
★ 77.5k🤯 LobeHub is your Chief Agent Operator, organizing your agents into 7×24 operations by hiring, scheduling, and reporting on your entire AI team.
Jeecgboot
★ 46.4kAI 低代码平台，「低代码 + 零代码」双模式驱动：低代码一键生成前后端代码，零代码 5 分钟搭建系统，AI Skills 一句话画流程、设计表单、生成整套系统。内置 AI聊天、知识库、流程编排、MCP插件等，兼容主流大模型。引领「AI 生成 → 在线配置 → 代码生成 → 手工合并->AI修改」开发模式，消除 Java 项目 80% 的重复工作，提效而不失灵活。
CowAgent
★ 44.7kCowAgent (chatgpt-on-wechat) 是基于大模型的超级AI助理，能主动思考和任务规划、访问操作系统和外部资源、创造和执行Skills、通过长期记忆和知识库不断成长，比OpenClaw更轻量和便捷。同时支持微信、飞书、钉钉、企微、QQ、公众号、网页等接入，可选择DeepSeek/OpenAI/Claude/Gemini/ MiniMax/Qwen/GLM/LinkAI，能处理文本、语音、图片和文件，可快速搭建个人AI助理和企业数字员工。
Minds Platform
★ 39.2kPlatform dedicated to building an open foundation for applied Artificial Intelligence, designed for people seeking production-ready AI systems they can truly control, extend and deploy anywhere.
Astrbot
★ 32.8kAI Agent Assistant & development framework that integrates lots of IM platforms, LLMs, plugins and AI feature, and can be your openclaw alternative. ✨
Browse More Business Applications MCP Servers
Explore all business applications servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up FastGPT in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use FastGPT?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.