🤖 ZBOT — ROZ AI Assistant

Your personal AI mascot that runs 100% locally. No cloud, no data leaving your machine.

What It Does

• Auto-detects your hardware (GPU, VRAM, RAM, CPU)
• Picks the best AI model your system can run
• Downloads and runs it locally via Ollama
• Can execute system commands (with your permission)
• Fully private — everything stays on your PC

Model Selection

Your Hardware	Model Selected
16GB+ VRAM, 32GB+ RAM	Llama 3.1 70B
8GB+ VRAM, 16GB+ RAM	Llama 3.1 8B
4GB+ VRAM, 8GB+ RAM	Llama 3.2 3B
2GB+ VRAM, 4GB+ RAM	Llama 3.2 1B
Anything else	TinyLlama

Install

# Install Ollama first
curl -fsSL https://ollama.com/install.sh | sh

# Clone and run
git clone https://github.com/zanicool/zbot.git
cd zbot
python3 zbot.py

Usage

You > what's my disk usage?
ZBOT > Let me check that for you.
⚡ Run 'df -h'? [y/n] y

Type exit to quit.

Requirements

• Linux
• Python 3.6+
• Ollama
• NVIDIA/AMD GPU recommended (CPU-only works but slower)

License

MIT