Jan

Founded by Ngoc Nguyen in 2023

Run open-source AI models locally or connect to cloud AI services

Cost

Free

Rating

★ People love it

Time to value

Quick Setup (< 1 hour)

You can use Jan to chat with AI models entirely on your own computer without sending data to external servers, or connect to cloud-based models like GPT-4, Claude, or Gemini. It works as a local alternative to ChatGPT, letting you download and run open-source models like Llama, Mistral, or DeepSeek. You can switch between local and online models, manage your conversation history, and customize how the AI responds based on your preferences. It's free, open-source, and works offline.

What Jan does

Download and run open-source LLMs locally on your computerConnect to cloud AI APIs like OpenAI or Anthropic from one interfaceChat with AI models without sending data to external serversSwitch between different AI models mid-conversationSave conversation history and user context across sessionsBrowse and install models from HuggingFace inside the appConfigure AI response style and preferences per user profileRun AI models fully offline on your own machine with no data sent to external serversSwitch between local open-source models and cloud models like GPT-4 or Claude from one interfaceDownload models directly from HuggingFace within the appStores conversation memory and user preferences so you don't repeat contextCompletely free and open-source with source code available on GitHubSupports many popular models including Llama, Mistral, DeepSeek, Qwen, and GemmaWorks on Mac and other operating systems with a desktop application