May we use cookies to learn which of our ads brought you here? Nothing is shared until you say yes. How we use them
Workspaces, databases, domains, an app store and a connection for your AI agents, on one dedicated machine. No Kubernetes, no per-seat fee, and serving 165 seconds after you click.
How the trial works: sign in, tell us about your institution, and once we approve it you get a dedicated 4 vCPU / 8 GB server under your own brand, free for 30 days. Afterwards the same server carries on as the Nano plan at €30.00 a month, and you can move to a bigger machine whenever you want. Want us to set it up for your team? Done-for-you setup →
Cloud Infrastructure
From hourly cloud machines to bare metal and on-demand GPU, the same platform runs on each. Pick the size the work needs, and change it when the work does.
(Flexible Development)
Plans from Nano to Titan: one term paid up front into your wallet (a month, or 6 or 10 months at a discount), then billed by the hour. Move up when the work grows and back down when it ends.
(High-Performance Apps)
A whole machine for the team with its own public IP — up to 32 vCPUs and 128 GB of memory on Titan, and more on request.
(Full Control & Isolation)
Our bare metal or your own server, with root access and a public address that’s yours down to the port.
(Custom GPU Needs)
GPU is on-demand — attached to a workspace when a job needs it, and released when it does not.
AI Infrastructure
Connect the AI agents your team already uses. They deploy, route and ship on the same platform, under the same rules as the person who connected them.
Protocol
Claude Code, Codex or any MCP client connects once, and works with the permissions of the engineer who connected it.
Guidance
Before its first move the agent is told the order the decisions happen in, so it does not guess at your stack.
Storage
Files, databases and running services stay between sessions, so the next prompt starts where the last one finished.
Audit
Each call an agent makes, allowed or refused, is written to the account and the client that made it.
Performance
Plans from Nano to Titan, each a machine with its own public IP. Move up as the work grows, and add GPU when a job needs it.
CPU Cores
32 vCPUs
On the Titan plan
Max RAM Capacity
128 GB
On the Titan plan
Disk
705 GB
On the Titan plan
Public IP
Dedicated
On every plan
Hourly Plans
Nano to Titan.
Billing
One term up front (1, 6 or 10 months), then hourly.
GPU Capacity
On-demand GPU support
Everything Included
Real Linux with root, SSH, VS Code in the browser and a desktop when you need one. It opens in a browser tab, so a cheap laptop is enough — no VM installs, no workstation per person.
MySQL, MariaDB, PostgreSQL, MongoDB, RabbitMQ and Valkey (Redis-compatible), managed, with per-user credentials. Nothing to install and nothing to keep running.
A public URL for every workspace, custom domains with automatic certificates, health checks, and deploy history you can roll back.
114+ templates to start from, n8n and Node-RED among them. Fork one, build it with logs, then publish it back so everyone after you starts where you finished.
A WireGuard address per person: reachable from anywhere, exposed nowhere. Nothing faces the internet except the routes you published.
Departments with enforced ceilings, your own identity provider — GitLab, Google, Entra or any OIDC issuer — and every action by a person or an agent on record. No per-seat fee.
Startup Impact
Less time building and babysitting machines, more time on the product. The platform is already set up, so the first day is spent on the work.
Deploy from a template, attach a domain and a database, and the work is on a real URL before the day is out.
A new engineer opens a browser tab and the team’s stack is already running in their workspace. A cheap laptop is enough.
Setup, routing and reading the logs are jobs an agent can do here, with the engineer’s own permissions.
You pay for the machine, not for the people on it. A fifth engineer or a tester does not change the bill.
Move up a plan, add a department, or bring your own server. The workspaces and templates come with you.
How It Works
Pick a plan, and 165 seconds later your platform is serving on its own machine with its own public IP.
Add people to your organization and give each team a department with its own limits. No per-seat fee.
Start from one of 114+ in the store, or build your own in the browser and publish it for everyone.
Real Linux with SSH, VS Code in the browser and a VPN address, opened from any laptop.
Every workspace gets a public address and a certificate. Add your own domain and a database beside it.
Point Claude Code or Codex at the platform. It works with the engineer’s role, and every call is recorded.
Owners and admins decide who joins, which department they sit in, and how much each department may use.
SSH, the editor in the browser, or an agent over MCP — everyone reaches the same workspace, under the same rules.
Documentation for every part of the platform, and a real person at the support address when you get stuck.
