N.O.M.A.D. itself is lightweight — it’s the tools and content you choose to install that determine the hardware you actually need. A barebones installation of the Command Center runs on modest hardware, but getting the most out of the AI tools requires a capable, ideally GPU-backed machine.
Project N.O.M.A.D. is not sponsored by any hardware manufacturer and is designed to be hardware-agnostic. Hardware listed here is for comparison purposes only.
Minimum specs
For a barebones installation of the Command Center and non-AI tools:
| Component | Minimum requirement |
|---|
| Processor | 2 GHz dual-core or better |
| RAM | 4 GB system memory |
| Storage | 5 GB free disk space |
| OS | Debian-based (Ubuntu recommended) |
| Internet | Required during installation only |
Optimal specs
To run LLMs and other AI tools at full performance:
| Component | Recommended |
|---|
| Processor | AMD Ryzen 7 or Intel Core i7 or better |
| RAM | 32 GB system memory |
| Graphics | NVIDIA RTX 3060 or AMD equivalent or better |
| Storage | 250 GB free disk space, preferably on SSD |
| OS | Debian-based (Ubuntu recommended) |
| Internet | Required during installation only |
More GPU VRAM allows you to run larger AI models. The community sweet spot is an AMD Ryzen 7 with 16 GB+ RAM. For detailed build recommendations at three price points (150–1,000+), see the Hardware Guide. Storage requirements
Storage needs scale with the content you download. Plan accordingly:
| Content | Approximate size |
|---|
| Full Wikipedia | ~95 GB |
| Khan Academy courses | ~50 GB |
| Medical references | ~500 MB |
| Maps (per US state) | ~2–3 GB |
| AI models | 10–40 GB depending on model |
Use an SSD for your primary storage. Disk speed directly affects AI response times and content search performance.
GPU acceleration for AI
Running AI locally requires significant computing power. Adding an NVIDIA GPU with the NVIDIA Container Toolkit can improve AI response speeds by 10–20x or more compared to CPU-only operation.
- CPU-only systems typically achieve 10–15 tokens per second
- GPU-accelerated systems typically achieve 100+ tokens per second
N.O.M.A.D. automatically detects NVIDIA GPUs when the Container Toolkit is installed. If you add a GPU after initial setup, go to Settings → Apps, find the AI Assistant, and click Force Reinstall to reconfigure the AI container with GPU support.
AMD GPUs are supported for general use. NVIDIA GPUs with the NVIDIA Container Toolkit provide the best AI acceleration support in the current release.
Further reading
For curated build recommendations at three price points — from a budget 150builduptoa1,000+ powerhouse — see the full hardware guide:
Hardware Guide → projectnomad.us/hardware 