​
QuantaCloud

QuantaCloud gives you on-demand NVIDIA GPU instances from one console: pick a GPU, pick a template, deploy, and connect over SSH or in your browser. You pay from a prepaid credit balance, with no commitments: the first hour is charged when you deploy, and the unused seconds of the current hour are refunded when you stop.

Quick Start

Go from sign-up to a running GPU in five steps.

Browse GPUs

Compare configurations and hourly prices.

Add Credits

Fund your account and start deploying.

Templates

Bare Metal, PyTorch + Jupyter, Open WebUI + Ollama, ComfyUI.

​
What you get

  • A single catalog — Every available GPU configuration, sorted by price, with real-time stock.
  • Self-serve deploy — Select a GPU, pick a template, and launch. SSH access is ready when your instance boots.
  • Browser-ready apps — JupyterLab, Open WebUI, and ComfyUI templates open at a private URL behind your QuantaCloud login.
  • Prepaid billing — Load credits, then launch. The first hour is charged when you deploy, each further hour when the previous one is used up, and unused seconds are refunded when you stop.
  • Managed SSH keys — A default SSH key is generated for you at signup. Bring your own keys too.
  • An API — Deploy instances, manage SSH keys, and check your balance from your own scripts. See the API Reference.

​
What you’ll need

  • A modern web browser
  • An SSH client (built into macOS and Linux; use PuTTY or Windows Terminal on Windows)
  • A credit or debit card for adding credits

​
Available GPUs

Instances come with 1, 2, 4, or 8 GPUs depending on the model. Availability changes in real time; the marketplace shows current stock and pricing.

​
Regions

All instances currently run in the United States, in Virginia and Midwest datacenters. See Regions for details.

​
How pricing works

All prices are per hour. The marketplace shows the hourly rate of each configuration, and disk is included with every instance. There are no egress or ingress charges. The first hour is charged from your credit balance when you deploy, and each further hour when the previous one is used up. When you stop, the instance is terminated, its disk is deleted, and the unused seconds of the current hour are refunded. If your balance can’t cover the next hour, the instance is terminated and its disk deleted. See Billing for details. There are no minimum commitments and no contracts. Need reserved or dedicated capacity? See quantacloud.net.