Installation Guide

Get ALICE running on your system

System Requirements

Recommended (NVIDIA GPU)

• GPU: NVIDIA RTX 3060 or better (12GB+ VRAM recommended)
• CUDA: 12.4 or later
• RAM: 16GB+ system RAM
• Storage: 50GB+ for models and images
• OS: Linux (Ubuntu 20.04+, Debian 11+) or macOS 12+

Also Supported

• AMD GPUs: ROCm 6.2+ compatible (RX 6000/7000 series)
• Apple Silicon: M1/M2/M3 with MPS acceleration
• CPU Only: Slower but functional (8+ cores recommended)

Quick Installation (Development)

# Clone the repository
git clone https://github.com/SyntheticAutonomicMind/ALICE.git
cd ALICE

# Create virtual environment
python3 -m venv venv
source venv/bin/activate

# Install PyTorch (choose your platform)
# NVIDIA CUDA:
pip install torch==2.6.0 torchvision==0.21.0 --index-url https://download.pytorch.org/whl/cu124

# AMD ROCm:
pip install torch==2.6.0 torchvision==0.21.0 --index-url https://download.pytorch.org/whl/rocm6.2

# Apple Silicon (macOS):
pip install torch==2.6.0 torchvision==0.21.0

# Install remaining dependencies
pip install -r requirements.txt

# Create directories
mkdir -p models images logs

# Start ALICE
python -m src.main

# Open in browser
open http://localhost:8080/web/

Production Deployment (Linux)

# Run installer (requires sudo)
sudo ./scripts/install.sh

# Add models to /var/lib/alice/models

# Start service
sudo systemctl start alice

# Enable auto-start
sudo systemctl enable alice

# Check status
sudo systemctl status alice

Production Deployment (SteamOS/Steam Deck)

# Run automated installer
./scripts/install_steamos.sh

# Add models to ~/.local/share/alice/models

# Check status
systemctl --user status alice

# View logs
journalctl --user -u alice -f

Model Setup

ALICE needs at least one Stable Diffusion model to generate images. Download from:

• CivitAI - Community models
• HuggingFace - Official models

Place models in the models/ directory. ALICE auto-detects model types (SD 1.5, SDXL, FLUX).

Verification

# Check GPU is detected
python -c "import torch; print(torch.cuda.is_available())"

# Test generation
curl -X POST http://localhost:8080/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{"messages":[{"role":"user","content":"a cat"}]}'

Next Steps

• Getting Started - Generate your first image
• Model Management - Download and organize models
• Web Interface - Using the dashboard