System requirements & workstation specs

The toolkit itself is light — the one thing that drives a bigger machine is running the AI LOCALLY (Ollama). Cloud AI, or a model on your own server, keep the workstation at the minimum.

Minimum: runs every tool with cloud AI or no AI — 64-bit dual-core, 8 GB RAM, ~500 MB free disk, no GPU, a 1280x800 or larger display.

Recommended: for daily use and local AI — quad-core, 16 GB RAM (32 GB to run a model locally), SSD 2 GB+ (plus 5-10 GB per local model), a GPU with 8 GB+ VRAM or Apple Silicon — only for local AI.

Local AI (Ollama): a small 7-8B model is the floor; 32 GB+ runs a capable 14-32B model and a Posse sweep without swapping. Cloud AI (Anthropic/OpenAI/Gemini with your key) needs none of this — just internet and a key.

Your machine: this page shows a live readout of this box (OS, CPU, RAM) with a color-coded local-AI readiness verdict, so you can see at a glance where you land against the tiers above.

Operating systems: Windows 10 version 1809 (build 17763) or later, or Windows 11 (64-bit); x64 desktop Linux with glibc 2.27 or newer (for example Ubuntu 22.04/24.04, Debian 12/13, Fedora 40+); or macOS 14 Sonoma or later. The app bundles its own runtime — nothing else to install.

What needs internet: the free tier needs none. Pro needs an online sign-in (a free account), then works offline for up to 30 days at a time — open Account while online at least once every 30 days to renew it. A trial has no offline grace: it needs that check at least every 14 days. Cloud AI and the internet tools (speed test, WHOIS, public IP, IP info, and NTP check against its default public pool.ntp.org) need internet whenever you use them. The Linux download's README has the full per-OS setup, admin rights, optional dependencies, and ports.