Running StaffGPT on your own computer
No technical background needed. This guide walks you through what a local setup involves, what you need, and roughly what it costs — in plain language.
Four simple steps
From your first call to a working AI team on your own machine.
1. We check your computer
You run a small one-click check (or we do it with you on a call). It tells us whether your computer is ready, or what to add.
2. We install everything
Our team sets up the StaffGPT software and downloads the AI model onto your machine. You don't touch a command line — we handle it remotely or on-site.
3. You pair the machine
You enter a short one-time code to connect your computer to your StaffGPT account. This links the two securely; nothing runs on its own until you say so.
4. Your AI team goes to work
Hire employees, pick which run locally, and they start working — with your data staying on your own computer.
Requirements
A simple checklist. If you're not sure, our setup check confirms all of this for you.
A reasonably modern computer
A desktop or laptop from the last few years. For the best experience, a machine with a dedicated graphics card (GPU) or a recent Apple Silicon Mac.
Enough memory and storage
At least 32 GB of memory (RAM) and about 40 GB of free disk space for the software and the AI model.
A stable internet connection
Needed once to download everything and to sync results. Day to day, the AI runs on your machine even if the internet drops.
Someone to press start
That's it. No IT department required — our team does the technical setup and shows you how to use it.
The AI model we install
Muse-Glimmer-30B-GGUF
This is the open-weight model that powers your local AI team. It's free to download, runs entirely offline once installed, and is the same model whether you run in the cloud or locally — so results stay consistent.
Not sure if your computer qualifies? Our free setup check confirms it in under a minute, or an engineer will verify it with you on a call.
What a local setup costs
An example budget for one workstation running a full local AI team. Real figures vary by hardware and how many machines you run — these are typical estimates.
One-time costs
- Workstation (recommended spec)$2,900 one-time
e.g. a desktop with a 24 GB GPU, or an Apple Silicon Mac with 32 GB. Skip this if you already own a suitable machine.
- Professional installation & setup$499 one-time
Our team installs the software, downloads and tunes the model, and pairs the machine with you.
- AI model (Muse-Glimmer-30B-GGUF)$0 included
Open-weight model — free to download. No per-use or licensing fee.
Ongoing costs
- StaffGPT Local plan$149 per month
Software updates, model refreshes, cloud sync, and support. Covers unlimited local employees on your machines.
- Electricity~$15 per month
Rough estimate for a workstation running during business hours. Varies by local rates and usage.
Example figures for planning only, in USD, and not a quote. Already own a capable computer? Your one-time cost drops to about $499. Contact sales for a tailored estimate.
Want us to handle the whole thing?
Book a setup call and an engineer will check your computer and get you running end to end.