Repolode
Submit
janhq

janhq/jan

Jan is an open-source desktop app for downloading and running LLMs locally, offering a ChatGPT-like interface with an optional cloud model connection.

Open-source alternative to:ChatGPT

View on GitHubWebsite
Stars
45,023
Forks
3,104
Open issues
592
Last commit
3 hours ago

What is it?

Jan is a desktop application that lets you download and run large language models on your own computer, with a chat interface similar to ChatGPT. Models such as Llama, Gemma and Qwen can be pulled from HuggingFace and run locally, so conversations can stay on your machine.

Jan can also connect to cloud model providers such as OpenAI, Anthropic, Mistral and Groq when you want them. It includes custom assistants for task-specific prompts, and runs a local OpenAI-compatible API server at localhost:1337 that other applications can call. Model Context Protocol (MCP) support is included for agent-style tool use.

Installers are available for Windows, macOS and Linux, including Microsoft Store and Flathub. The project is built with Tauri and Rust, and the README includes instructions for building from source with Make or Yarn.

Key features

  • 01Download and run open models such as Llama, Gemma and Qwen from HuggingFace.
  • 02Connect to cloud providers including OpenAI, Anthropic, Mistral and Groq.
  • 03Create custom assistants for specific tasks.
  • 04Expose a local OpenAI-compatible API at localhost:1337.
  • 05Support Model Context Protocol (MCP) for agent-style tool use.

Use cases

  • Chatting with an LLM offline while keeping prompts and data on your own machine.
  • Serving a local OpenAI-compatible endpoint for other tools during development.
  • Comparing open-weight models without sending requests to a cloud provider.

Quick start

  1. 1

    Download the app for your OS from jan.ai or GitHub Releases

  2. 2

    Clone the repository

    git clone https://github.com/janhq/jan
  3. 3

    Enter the project folder

    cd jan
  4. 4

    Install dependencies, build and launch the app with Make

    make dev

Things to know

  • Building from source requires Node.js, Yarn, Make and Rust to be installed first.
  • Building on Windows needs Visual Studio Build Tools, LLVM, Ninja and CMake, and Make must run from Git Bash.
  • Local models need substantial RAM, and GPU acceleration depends on your hardware.

Topics

#llm#local-ai#chatgpt-alternative#desktop-app#openai-compatible#offline#tauri#rust#chatgpt#gpt

FAQ

What is jan?
Jan is an open-source desktop app for downloading and running LLMs locally, offering a ChatGPT-like interface with an optional cloud model connection. Jan is a desktop application that lets you download and run large language models on your own computer, with a chat interface similar to ChatGPT. Models such as Llama, Gemma and Qwen can be pulled from HuggingFace and run locally, so conversations can stay on your machine.
Is jan free to use? What license does it have?
jan is open source under the Other license. Check the license terms before using it in commercial products.
Is jan still maintained?
Actively maintained. The last commit to jan was on Oct 21, 2026.
What can I use jan for?
Chatting with an LLM offline while keeping prompts and data on your own machine. Serving a local OpenAI-compatible endpoint for other tools during development. Comparing open-weight models without sending requests to a cloud provider.
How do I install jan?
Download the app for your OS from jan.ai or GitHub Releases Clone the repository: git clone https://github.com/janhq/jan Enter the project folder: cd jan Install dependencies, build and launch the app with Make: make dev
Where can I get jan?
The source code and installation instructions are on GitHub: https://github.com/janhq/jan

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects