Jan AI
jan.ai
Jan is an open source alternative to ChatGPT that runs 100% offline on your computer
Description
Jan is an open-source alternative to ChatGPT. Run open-source AI models locally or connect to cloud models like GPT, Claude and others. · An open model ecosystem · Over 4 million downloads · All the tools you need to make Jan yours
Jan AI is an open-source platform that allows users to run AI language models directly on their personal computers. It prioritizes privacy and local processing, enabling users to interact with AI models without an internet connection or sharing data externally. Jan AI provides an OpenAI-compatible API server for integration with other applications
Installation
| Installs into | macos, windows, linux |
Not sure where to start — ask an assistant to walk you through:
Features
LOCAL MODEL EXECUTION, PRIVACY-FOCUSED PROCESSING, OPENAI-COMPATIBLE API SERVER, CUSTOMIZABLE EXPERIENCE, MULTI-PLATFORM SUPPORT, OPEN-SOURCE DEVELOPMENT
Use cases
PERSONAL AI ASSISTANT, OFFLINE AI PROCESSING, PRIVATE DATA ANALYSIS, CUSTOM AI MODEL DEPLOYMENT, LOCAL LANGUAGE PROCESSING
FAQ
Jan AI is an open-source, privacy-focused AI chat assistant that runs 100% offline on desktop (Windows, Mac, Linux) and soon mobile devices. It offers an alternative to cloud-based AI models like ChatGPT by allowing you to use local or cloud AI models with full control, privacy, and no mandatory internet connection.
Jan AI's key features include running fully offline for local AI models, ensuring privacy; supporting both local open-source models (e.g., Llama3, Mistral, Qwen) and cloud-based AI providers (OpenAI, Anthropic, etc.); offering customizable assistant instructions and model parameters; featuring an extensible architecture with plugin and API support; providing Retrieval-Augmented Generation (RAG) for document analysis; and utilizing GPU acceleration for performance optimization.
Jan-v1 is a 4 billion parameter AI model integrated into Jan AI, which operates fully offline with 91.1% accuracy on question-answering tasks, surpassing many other models. It is lightweight and designed for quick, low-latency responses without needing the cloud.
Yes, Jan AI can run local models entirely offline, providing complete data privacy. However, certain advanced features like integration with external APIs or cloud models require internet access.
To get started with Jan AI, you need to download and install Jan from jan.ai, then download or add an AI model locally or connect to a cloud API key. After that, launch Jan, select your model, and start chatting. Detailed setup is available in the Quickstart guide on their documentation page.
Yes, you can optionally connect Jan to cloud AI providers by adding your API keys, enabling you to switch seamlessly between local and cloud models within the same interface.
The Jan desktop application and running local AI models are free and open-source. Cloud model usage depends on the external provider's pricing and API credits.
Since Jan runs local models on your device without uploading data to the cloud, all conversation data remains private on your machine. You have full control over which models run and how data is handled.
Jan includes tools to help select models compatible with your computer hardware. It supports GPU acceleration when available to improve performance automatically.
Support is available through GitHub issue tracking, Discord communities, and email contacts provided on the Jan website.
Specs
| Type | Agent |
| Section | Productivity / Local LLM Runner |
| Pricing | free (от $0/mo) |
| Platform | Web + desktop |
| Systems | macos, windows, linux, api, web |
| Installs into | macos, windows, linux |
| Who for | Individual |
| Site language | en |
| GitHub | menloresearch/jan |
| Rating | 4.00 (1 reviews) |
| Views | 356 912 |
| Launched | 2024-09-04 |
