Jan is an open-source ChatGPT alternative that runs 100% offline on your computer. With a beautiful, intuitive interface and support for multiple AI models, Jan makes local AI accessible to everyone. It stores all data locally, ensuring complete privacy while providing a smooth, familiar chat experience similar to commercial AI assistants.
📑 Table of Contents
Key Features
- 100% Offline – No internet required after setup
- Privacy First – All conversations stay on your device
- Beautiful UI – Modern, clean chat interface
- Model Hub – Easy download of popular open-source models
- Extensions – Plugin system for added functionality
- Local API Server – OpenAI-compatible REST API
- Cross-Platform – Linux, Windows, macOS support
System Requirements
- RAM – 8GB minimum, 16GB recommended
- Storage – 5GB+ for application and models
- GPU – Optional: NVIDIA GPU for faster inference
- CPU – Modern multi-core processor
Installation on Linux
# Download the AppImage
wget https://github.com/janhq/jan/releases/latest/download/jan-linux-x86_64.AppImage
# Make it executable
chmod +x jan-linux-x86_64.AppImage
# Run Jan
./jan-linux-x86_64.AppImage
# Or install via Snap
snap install jan
Getting Started
- Launch Jan and go to the Hub section
- Browse available models (Llama, Mistral, Phi, etc.)
- Click Download on your preferred model
- Start chatting once the download completes
Recommended Models
- Mistral 7B – Best balance of speed and quality
- Llama 2 7B – Meta’s versatile open model
- Phi-2 – Microsoft’s compact efficient model
- TinyLlama – For systems with limited resources
- CodeLlama – Specialized for programming tasks
Local API Server
Jan can serve as a local API endpoint:
# Enable the API server in Settings
# Default endpoint: http://localhost:1337/v1
# Use with any OpenAI-compatible client
from openai import OpenAI
client = OpenAI(base_url="http://localhost:1337/v1", api_key="jan")
response = client.chat.completions.create(
model="mistral-7b",
messages=[{"role": "user", "content": "Hello!"}]
)
Features
- Conversation History – All chats saved locally
- Custom Instructions – Set system prompts per conversation
- Model Switching – Change models mid-conversation
- Temperature Control – Adjust creativity levels
- Export/Import – Backup and restore conversations
Use Cases
- Private AI Assistant – ChatGPT-like experience locally
- Writing Help – Draft emails, documents, content
- Code Assistance – Programming help and explanations
- Learning – Study aid and research assistant
- Brainstorming – Idea generation and exploration
Was this article helpful?