😎 Private AI with no more guesswork.
Tell us which models, users, agents, and workloads you need. We translate those requirements into a complete server specification and step-by-step guided deployment path — then you run one installation script on infrastructure you control.
Three supported deployment paths
Customers choose a supported deployment location. Terminal.Glass remains the product they buy and use. We remove the CPU, memory, storage, GPU, operating-system, and cloud-instance guesswork.
NoCloudGPT — Ubuntu Linux VMs and standalone servers
Run Terminal.Glass on an Ubuntu Linux 24/26 virtual machine or standalone server you control — customer-owned, customer-controlled private AI on local, virtualized, or bare-metal infrastructure.
- Ubuntu Linux 24/26 VM or standalone server
- Completely private when inference stays on your hardware
- One installation script, one Terminal.Glass instance
AWS — customer-owned account
Terminal.Glass recommends the appropriate AWS instance, matched to selected models and workloads, with one-script deployment in a cloud account you own.
- Guided instance sizing for your workload
- Customer owns the AWS account
- Complete guided deployment instructions
DigitalOcean — customer-owned account
Terminal.Glass recommends the appropriate DigitalOcean server, matched to selected models and workloads, with one-script deployment in an account you control.
- Guided server sizing for your workload
- Customer owns the DigitalOcean account
- One script, one Terminal.Glass instance
Coming soon
Google Cloud Platform and Microsoft Azure support are in development. Contact us if you are planning a deployment on GCP or Azure.
- Google Cloud Platform — coming soon
- Microsoft Azure — coming soon
From requirements to running system
We handle the server specification and remove the guesswork. You create the recommended server, run one installation script, and receive one complete Terminal.Glass instance.
Share your requirements
Models, users, agents, workloads, and preferred deployment path — NoCloudGPT, AWS, or DigitalOcean.
Receive your server specification
CPU, RAM, storage, GPU, operating system, and provider instance — sized for your deployment.
Run one installation script
😎 Complete guided deployment instructions on infrastructure you control.
Your team starts
Happy Nerds Menu, built-in DokuWiki, Glass Agents, and Jet Agents on one Terminal.Glass instance.
Your server. Your data. Your control.
The customer owns or controls the server or cloud account. Terminal.Glass does not need access to your server by default.
One server license activates one Terminal.Glass instance on one computer or cloud server. Deployment documentation and available support options are provided with your account.