Frequently asked questions

An hourly GPU rental service for customers in the EU. Developers, students and companies rent graphics cards for AI, rendering and compute, or launch ready-made apps for images, video and chat. Prices are in euros and you pay by the minute.

The catalogue lists machines with graphics cards in data centres. You choose what to run: a ready-made app for images, video or chat, or a developer server with PyTorch and Jupyter. Everything runs in an isolated container with SSH and Jupyter access, billed by the minute. The server disk is wiped when you stop, but you can save your working folder and get it back on your next server.

We don’t build our own data centres. We bring available machines from several providers into one catalogue, so an hour costs less than at clouds that run their own facilities.

Yes: dedicated machines for long projects and multi-GPU configurations. Write to us through the contact form and we will discuss the terms.

20 models, from the RTX 3060 to the H100, H200 and B200, including the RTX 3090, 4090 and 5090, the professional RTX A series and the data-centre L4, L40, L40S and A100. The current list with prices is on the pricing page.

In data centres run by our providers. The country of every machine is shown in the catalogue before you launch, so you can choose where your workload runs.

Yes. Create a token in the dashboard under Settings → API tokens. With the API you can find an available machine, launch a server with your own Docker image, check its status and SSH details, and stop it. API documentation →

Sign up, confirm your email and top up your balance. Then click Rent a GPU in the dashboard, choose what to run, then the card and the machine, and click Launch. Your balance must cover one hour of the chosen machine. Connection details appear in the dashboard and arrive by email.

From €0.12 an hour for an RTX 3060, €0.30 for an RTX 3090, €0.59 for an RTX 4090, €1.46 for an A100 and €3.90 for an H100. Prices are per hour for a machine with one card, billed by the minute. Current prices for every model are in the server catalogue.

Every minute a server runs, 1/60 of its hourly price is deducted from your balance. The minimum charge per launch is 15 minutes. To launch, your balance must cover at least one hour of the server; that amount is not charged up front. Internet traffic is billed separately, as used. If your balance runs out, the server stops and we email you.

Yes, per gigabyte downloaded and uploaded. The rate depends on the machine, is usually around a cent per GB or less (B200 and H200 can cost more) and is shown before you launch. We favour machines with cheap traffic.

No long commitment: you pay by the minute, with a minimum charge of 15 minutes per launch. Launch a machine, test your code and stop it.

Usually 2–5 minutes, sometimes up to 10, while the machine downloads the image and starts the container. With your own Docker image, 8–15 minutes. You are not charged while the server is getting ready.

With the SSH command from the dashboard, or through Jupyter Lab in your browser. Upload your SSH key in the settings to make this easier.

Ready-made apps install together with their models and open in your browser: Images (Flux, Qwen-Image, HiDream-O1, SDXL), Photo editor (Qwen Edit), Video (Wan 2.2 14B and 5B), Private ChatGPT (Qwen3 32B and 8B, MiMo 9B) and Text to speech (Qwen3-TTS, VoxCPM2, Chatterbox). For developers there is PyTorch with Jupyter, vLLM with an OpenAI-compatible API, an empty ComfyUI and your own public Docker image: specify it at launch and we add SSH.

Yes. Use any public image from Docker Hub or the GitHub Container Registry.

Yes, 2, 4 and 8 GPU configurations are available for distributed training and large models.

The server disk is wiped when you stop. If you need your models and results, click Save and stop: we keep the /workspace folder for €0.03 per GB per month, and your next server gets it automatically. Saving does not work yet for custom Docker images.

If a server has been idle for about two hours, we email you. If it is still idle an hour later, we save its folder and stop it, so you don’t pay for nothing. You can turn this off in the settings for long jobs.

You don’t pay for the downtime, and we help you restart the job on another machine. That is why it is worth saving intermediate results regularly.

Look at the video memory. 7–8B models in 4-bit run on cards from 12 GB. A 24 GB card (RTX 3090, 4090) fits models up to 36B in 4-bit, including 32B. 70B in 4-bit fits on one 48 GB card (RTX A6000, L40S) at normal context; for long context or many users take 80 GB (A100, H100).

Yes. The A100, H100, H200 and B200 with HBM memory and multi-GPU configurations are there for exactly that.

Yes: Blender Cycles, Octane, Redshift and Unreal run from the command line. For animation it is handy to run several machines in parallel.

In the Balance section of the dashboard, by card or SEPA transfer in euros, from €10. Payments go through a payment provider that we will announce before launch. Until then, top-ups may be unavailable.

Prices on the site are in euros. How VAT applies to your account and country will be shown at checkout, before you pay.

You get an email for every top-up, and the full history of top-ups and charges is in the dashboard. Proper invoices will be issued once our EU company is registered.

Yes. You can request a refund of your unspent paid balance in the dashboard; it goes back to the original payment method.

There is no pause: once you stop, the machine is released and billing stops. To keep your work, click Save and stop: we keep /workspace for €0.03 per GB per month (charged daily), and your next server gets it automatically. Saving does not work yet for custom Docker images.