Skip to content
New NVIDIA B200 now available Reserve capacity
reviosa cloud
Documentation

From sign-up to running GPU

Everything you need to operate Reviosa Cloud from the console: instances, regions, storage, billing, team access and keys.

Quickstart

You can go from zero to a running GPU in a few minutes. No sales call and no quota request are required for on-demand instances.

  1. Create an account at Sign up with your work email. New accounts receive free starter credit.
  2. Add a payment method or credit under Billing. Payments are processed by Stripe; card details never touch our servers.
  3. Launch an instance from Instances → Launch: choose a GPU type, a region and on-demand or reserved billing.
  4. Connect using the IP address shown on the instance page once its status turns running.
  5. Stop or terminate when you are done — compute billing ends at that second.
Tip: start on a smaller card (from $0.89/hr) to validate your environment, then change the instance type for the real run.

Instances

An instance is a GPU or CPU machine billed per second. Each instance moves through four states:

StateWhat it meansCompute billing
provisioningHardware is being allocated and imaged.Starts
runningReady to use.Yes
stoppedPowered off; attached volumes are kept.No
terminatedReleased permanently.No

Billing modes

  • On-demand — billed per second at the hourly rate. Best for experiments and bursts.
  • Reserved — billed at a lower committed monthly rate. Best for steady, long-running workloads.

See pricing for live rates, or read On-demand vs reserved to choose.

Regions

Pick the region closest to your users or to where your data must stay.

Region codeLocation
us-east-1US East (Ashburn, VA)
us-west-2US West (Portland, OR)
eu-central-1EU Central (Frankfurt)
ap-northeast-1Asia Pacific (Tokyo)

Choosing a region is a trade-off between latency, data residency and capacity — our region guide walks through it.

Volumes & storage

NVMe block volumes live independently of instances. Create a volume under Volumes, attach it to an instance in the same region, and detach it before terminating the instance to keep your data.

  • Volumes persist across stop/start and can move between instance types.
  • Use object storage for datasets and artifacts you share across regions or teams.
  • Checkpoint long training runs to a volume so a restart never loses more than one interval.

Billing & credits

  • Per-second metering with a 60-second minimum per instance start.
  • Credits: top up prepaid credit from Billing; usage draws from credit first.
  • Invoices and receipts are available as PDFs from the billing page.
  • Usage shows spend by day and by instance, plus a budget tracker so surprises show up early.

Team & access

Invite teammates by email from Team. Everyone works in the same account and shares instances, volumes and billing, so you never pass around a single login.

API keys

Create keys under API keys for automation. A key is shown once at creation and stored only as a hash — copy it into your secret manager immediately. Revoke keys you no longer use; revocation takes effect at once.

Account security

  • Enable two-factor authentication from Settings.
  • Review and sign out active sessions on other devices from the same page.
  • Read how we protect infrastructure and data on Trust & security.

Still stuck?

Engineers answer support tickets — not a script.

Open the support center
Get started

Start training in minutes

Create an account, add a card, and launch your first GPU instance. Per-second billing means you only pay for what you use.

No minimum commitment · Cancel anytime · $10 free credit for new accounts