Skip to content

Free games, apps and software, straight from official sources

RSS
Jan logo

Jan Download: An Offline ChatGPT Alternative for Your Desktop

Jan · for Windows, macOS and Linux

Developer: Jan (janhq) (opens in a new tab)

Open source
  • Windows
  • macOS
  • Linux
DownloadFrom the official source · Windows, macOS and Linux (opens in a new tab) Official website (opens in a new tab)

The download link points to the program’s official source; the file is not hosted on this site. Link address: github.com

Link not working? Let us know
What is the problem?

Your report goes only to the site’s editor; we never ask for your name or address.

Jan is a free, open-source desktop app that downloads and runs large language models entirely on your own computer, with optional connections to cloud providers. It runs on Windows, macOS and Linux and is licensed under Apache 2.0.

2 min read

Jan describes itself as an open-source replacement for ChatGPT. Instead of a terminal, you get a familiar chat window: pick a model, let Jan download it, and start talking. When you use local models, everything stays on your machine.

What does Jan do?

  • Local models: Browse and download open models such as Llama, Gemma, Qwen and gpt-oss from Hugging Face, then run them on your own hardware.
  • Optional cloud models: If you want, you can also connect it to providers such as OpenAI, Anthropic, Mistral and Groq and use them from the same window.
  • Custom assistants: Set up assistants with their own instructions for specific tasks.
  • Local API: A built-in OpenAI-compatible server at localhost:1337 lets other apps use the models running in Jan.
  • MCP support: Model Context Protocol integration gives models access to tools.

Who is it for?

Jan is a good fit if you want a private chatbot but would rather click than type commands. It also suits developers who want a desktop app and a local API in one package.

How to install

Use the Download button to open the project’s official releases page and pick the installer for your system:

  • Windows: jan.exe
  • macOS: jan.dmg (a universal build for Apple Silicon and Intel)
  • Linux (x64): a .deb package or an AppImage

Jan is also listed on the Microsoft Store and on Flathub (ai.jan.Jan). Developers can build it from source:

git clone https://github.com/janhq/jan
cd jan
make dev

Building from source needs Node.js 20 or newer, Yarn, Make and Rust. On Windows, run make from Git Bash rather than cmd.exe.

System requirements

  • Windows: Windows 10 or later. GPU acceleration works with NVIDIA, AMD and Intel Arc cards.
  • macOS: macOS 13.6 or later. For Macs the project lists 8 GB of RAM for 3B models, 16 GB for 7B models and 32 GB for 13B models.
  • Linux: Most distributions work, with GPU acceleration available.

License

Jan is free and open source under the Apache 2.0 license. The models you download have their own licenses.

Pros

  • Simple chat window, no command line needed
  • Runs local models fully offline
  • Built-in OpenAI-compatible API for other apps

Cons

  • Larger models need a lot of RAM
  • Speed depends heavily on your graphics card

Frequently Asked Questions

Does Jan need an internet connection?

Only to download the app and the models. After that, local models run completely offline on your computer.

How much RAM do I need for Jan?

It depends on the model size. For Macs the project lists 8 GB for 3B models, 16 GB for 7B models and 32 GB for 13B models; bigger models need more memory on any system.

Similar software

Show all: AI Tools

llama.cpp

Download llama.cpp 0.7.0, the free, MIT-licensed engine that runs open GGUF language models locally on CPU or GPU, with an OpenAI-compatible server. Install commands and first steps.

Open source

  • Windows
  • macOS
  • Linux
Version 0.7.0

llamafile

Download llamafile from Mozilla.ai: a single executable file that runs open language models locally with nothing to install. Quick start and tips.

Open source

  • Windows
  • macOS
  • Linux
Version 0.10.6

Open WebUI

Install Open WebUI 0.12.4, the free, self-hosted AI interface that connects to Ollama and OpenAI-compatible APIs, with document uploads, web search and multi-user support.

Free

  • Windows
  • macOS
  • Linux
Version 0.12.4

LocalAI

Install LocalAI, the free, MIT-licensed engine that runs text, image and audio models locally behind an OpenAI-compatible API. No GPU required.

Open source

  • Linux
  • macOS
  • Windows