Jan logo

Jan

Free tier

Personal AI intelligence that runs locally and answers only to you

Free·All audiences·Powered by Multi-provider (Llama, Mistral, Claude, GPT-4, Gemini, DeepSeek, Qwen, Gemma, and more)·API available·Open source

Key strengths

Fully open-source with 42.8K+ GitHub starsRuns AI models 100% locally for complete privacySupports both local open models and remote API-based models (OpenAI, Anthropic, Google, etc.)Persistent memory context across conversationsCross-platform desktop app with API reference for developers
Completely free
Singapore, Singapore
Self-hostable
No ratings yet

Developer Documentation

Installation & Setup

Download the latest release (e.g., v0.8.2) from the Jan GitHub releases page. Jan is available for macOS, Windows, and Linux. The repository is fully open-source under a permissive license.

Local API

Jan exposes a local API server that is compatible with OpenAI's API format, allowing you to point existing OpenAI SDK integrations at Jan's local endpoint with minimal code changes:

# Example: Query Jan's local API
curl http://localhost:1337/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "mistral-7b",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Model Integration

  • Local models: Browse and pull models from HuggingFace (123+ available) directly within the app or via config files.
  • Remote models: Configure API keys for OpenAI (ChatGPT), Anthropic (Claude), Google (Gemini), Mistral AI, DeepSeek, Moonshot AI (Kimi), Alibaba (Qwen), and Meta (Llama API).

Key Resources

  • API Reference: Available at jan.ai/api-reference
  • Docs: jan.ai/docs
  • Changelog: Track releases and feature updates at jan.ai/changelog
  • GitHub: Full source code, issues, and contribution guides at github.com/janhq/jan