jan
Jan is an open source alternative to ChatGPT that runs 100% offline on your computer.
The Jan MCP server enables developers to integrate Jan's offline AI capabilities into MCP-compatible applications. Jan is an open-source alternative to ChatGPT that runs large language models entirely on your local machine without internet connectivity. Through its MCP integration and OpenAI-compatible API running on localhost:1337, developers can build AI-powered tools that leverage local LLMs from HuggingFace while maintaining complete privacy and control. It supports both local inference and optional cloud model connections for flexible deployment scenarios.
Key Features
Use Cases
- 01Building privacy-conscious AI tools that never send data to external servers
- 02Developing MCP-enabled applications that run LLMs entirely offline
- 03Creating specialized AI assistants for domain-specific tasks on local hardware
- 04Integrating local LLM capabilities into existing applications via OpenAI-compatible API
- 05Prototyping AI features without cloud API costs or network dependencies
- 06Running AI workloads in air-gapped or restricted network environments
Related MCP Servers
View moremarkitdown
Python tool for converting files and office documents to Markdown.
firecrawl
The context API to search, scrape, and interact with the web at scale. 🔥
prompts.chat
f.k.a. Awesome ChatGPT Prompts. Share, discover, and collect prompts from the community. Free and open source — self-host for your organization with complete privacy.
langflow
Langflow is a powerful tool for building and deploying AI-powered agents and workflows.
jan — FAQ
What is the Jan MCP server?+
The Jan MCP server is an integration that connects Jan's offline AI platform with Model Context Protocol-compatible applications. It allows developers to use local LLMs running through Jan in their MCP-enabled tools while maintaining complete privacy and control.
How do I install Jan for MCP integration?+
Download the Jan desktop application for your platform from jan.ai or GitHub releases (jan.exe for Windows, jan.dmg for macOS, or jan.deb/AppImage for Linux). Once installed, Jan provides MCP capabilities and runs an OpenAI-compatible API server on localhost:1337 that MCP clients can connect to.
Which AI clients work with Jan's MCP server?+
Jan works with any MCP-compatible client through its Model Context Protocol integration. It also supports applications using OpenAI-compatible APIs via its local server endpoint.
Do I need API keys to use Jan?+
No API keys are required for local LLM execution. However, if you choose to connect cloud models (OpenAI GPT, Anthropic Claude, Mistral, Groq), you'll need API keys for those respective services.
Is Jan free and open source?+
Yes, Jan is released under the Apache 2.0 license and is completely free to use. All source code is available on GitHub and the application can be built from source.
What are the system requirements for running Jan?+
Minimum requirements are macOS 13.6+ or Windows 10+ or most Linux distributions. RAM needs vary by model size: 8GB for 3B parameter models, 16GB for 7B models, and 32GB for 13B models. GPU acceleration is available for NVIDIA, AMD, and Intel Arc on Windows and macOS.
How do I install jan?+
Open the source repository on GitHub and follow its README. jan is a mcp server — MCP Agents Market links you directly to the official repo.
Is jan free?+
jan is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.