GPU Cloud, Priced by the Hour

When you launch an instance, you are billed per hour based on the GPU type in your template. Billing starts when the instance enters 'running' state and stops when you stop or terminate the instance. Partial hours are billed by the minute.

Credits are Quest Labs's billing unit. You purchase credits upfront and they are consumed based on GPU hourly rates. Purchased credits never expire. Trial account credits are valid for 90 days.

You create a template by specifying a Docker image, startup script, and GPU type. After a quick review, the template is approved and you can launch instances from it. The platform handles all provisioning and orchestration.

Yes. Each instance includes persistent storage that survives stops and restarts. Your datasets, model checkpoints, and installed packages are preserved. Data is only lost when you explicitly terminate an instance. Free tier storage included with additional storage available.

Yes. You can submit any public or private Docker image as a template, including custom images from Docker Hub or private registries. Popular choices include PyTorch, TensorFlow, Hugging Face Transformers, and NVIDIA NGC containers. Templates are reviewed before deployment.

Starter ($25) is ideal for individual developers or experimentation. Pro ($100) is suited for regular AI development with more credits at a better rate. Team ($500) is designed for teams needing large-scale GPU compute with the best per-credit pricing.

We accept credit cards (Visa / Mastercard / JCB), Apple Pay, Google Pay, and USDT. Enterprise customers can request invoiced billing.

At this time, purchases are non-refundable. We recommend confirming your needs carefully before completing payment.

Quest Labs employs enterprise-grade security standards: all data in transit is TLS encrypted, data at rest uses AES-256 encryption, each instance runs in a fully isolated environment, and we maintain SOC 2 Type II certification with GDPR compliance.