Omarchy AI - Seamless Installation Guide

July 18, 2025 · View on GitHub

Single Command Installation

curl -fsSL https://raw.githubusercontent.com/mitkox/omarchy-ai/refs/heads/master/boot.sh | bash

This command will:

  • ✅ Download the latest Omarchy AI
  • ✅ Install base Omarchy system (Hyprland, Waybar, etc.)
  • ✅ Install AI development tools (Python, PyTorch, etc.)
  • ✅ Create conda environment for AI development
  • ✅ Setup Jupyter Lab and ML libraries
  • ✅ Configure GPU support and monitoring
  • ✅ Create AI workspace directory
  • ✅ Setup shell aliases and commands

Installation Time

  • Fresh system: 20-40 minutes
  • Existing Arch system: 10-20 minutes
  • Network dependent: Package downloads

What Gets Installed

Base Omarchy System

  • Hyprland (Wayland compositor)
  • Waybar (status bar)
  • Wofi (application launcher)
  • Alacritty (terminal)
  • Neovim (editor)
  • Theme management system

AI Development Environment

  • Python 3.11 in dedicated conda environment
  • PyTorch with CUDA support
  • Jupyter Lab for interactive development
  • Transformers (Hugging Face)
  • MLflow for experiment tracking
  • TensorBoard for visualization
  • Various ML libraries (numpy, pandas, scikit-learn, etc.)

AI Tools & Commands

  • ai-env - Activate AI development environment
  • ai-workspace - Navigate to AI workspace
  • jupyter-ai - Start Jupyter Lab
  • mlflow-ui - Start MLflow UI
  • tensorboard-ai - Start TensorBoard
  • gpu-monitor - Monitor GPU usage

Post-Installation Usage

1. Reboot (Required)

sudo reboot

2. Activate AI Environment

# Open a new terminal and run:
ai-env

3. Verify Installation

python --version    # Should show Python 3.11.x
pip list           # Shows installed packages
nvidia-smi         # Check GPU (if available)

4. Start Developing

ai-workspace       # Go to AI workspace directory
jupyter-ai         # Start Jupyter Lab

AI Workspace Structure

After installation, you'll have:

~/ai-workspace/
├── projects/      # Your AI projects
├── models/        # Downloaded models
├── datasets/      # Training data
├── experiments/   # ML experiments
├── notebooks/     # Jupyter notebooks
├── logs/          # Training logs
├── mlruns/        # MLflow tracking
└── .env          # Environment variables

Troubleshooting

If ai-env doesn't work:

# Method 1: Manual activation
source /opt/miniconda3/etc/profile.d/conda.sh
conda activate ai-dev

# Method 2: Fresh shell
bash --login
ai-env

# Method 3: Direct script
~/.local/share/omarchy-ai/activate-ai-env.sh

If conda is not found:

# Ensure conda is in PATH
echo 'export PATH="/opt/miniconda3/bin:$PATH"' >> ~/.bashrc
source ~/.bashrc

If packages are missing:

ai-env
pip install transformers torch torchvision torchaudio

Manual Installation (Advanced)

1. Clone Repository

git clone https://github.com/mitkox/omarchy-ai.git ~/.local/share/omarchy-ai
cd ~/.local/share/omarchy-ai

2. Run Installation

chmod +x install.sh
./install.sh

3. Follow post-installation steps above

Key Features

🔒 Privacy-First

  • Local AI inference with llama.cpp
  • Offline-first development
  • No cloud dependencies required

🚀 Performance

  • CUDA GPU acceleration
  • Optimized for AI workloads
  • Real-time monitoring

🛠️ Developer-Friendly

  • Pre-configured development environment
  • Integrated tools and libraries
  • Version control with Git LFS

📚 Complete Ecosystem

  • Jupyter Lab for interactive development
  • MLflow for experiment tracking
  • TensorBoard for visualization
  • Container support for deployment

System Requirements

Minimum

  • OS: Arch Linux (fresh installation recommended)
  • RAM: 8GB (16GB recommended for AI workloads)
  • Storage: 50GB free space
  • Network: Internet connection for installation
  • OS: Fresh Arch Linux installation
  • RAM: 32GB+ for large models
  • GPU: NVIDIA GPU with CUDA support
  • Storage: SSD with 100GB+ free space

Support

Getting Help

  • Check the troubleshooting section above
  • Review logs in ~/.local/share/omarchy-ai/logs/
  • Open an issue on GitHub

Common Issues

  • Installation fails: Check internet connection and disk space
  • GPU not detected: Install NVIDIA drivers first
  • Conda issues: Reboot and try again
  • Permission errors: Ensure user has sudo access

What's Next?

After successful installation:

  1. Download your first model:

    ai-env
pip install huggingface_hub
huggingface-cli download Qwen/Qwen3-0.6B

  2. Start your first project:

    ai-workspace
mkdir my-first-ai-project
cd my-first-ai-project
jupyter-ai

  3. Explore the ecosystem:

    • Try the pre-installed packages
    • Check out the example notebooks
    • Monitor your GPU usage
    • Set up experiment tracking

Welcome to Omarchy AI - your complete local AI development environment! 🚀