Wollnut Labs
All Posts
Product

Why Per-Minute Billing Changes Everything for GPU Cloud

Wollnut Labs TeamMarch 25, 20253 min

The Problem with Hourly Billing

Most GPU cloud providers bill by the hour with a one-hour minimum. This means:

  • A 10-minute test run costs you a full hour
  • Quick iterations during development waste credits
  • You're incentivized to batch work rather than iterate fast

    • Per-Minute Billing

    At Wollnut Labs, we bill by the minute. Our background billing engine tracks each running instance and charges your credit balance every 60 seconds.

    Example: You spin up an H100 to run a quick evaluation. It takes 12 minutes. You pay for 12 minutes ($0.45), not 1 hour ($2.25). That's an 80% savings on short tasks.

    How It Works

  • When you start an instance, our billing CRON begins tracking
  • Every minute, we calculate the cost and deduct from your balance
  • When you stop the instance, billing stops immediately
  • If your balance hits zero, instances auto-stop — no surprise charges

    • Auto-Stop Protection

    We've built multiple safety nets:

  • **Low balance warnings** when credits drop below $2
  • **Auto-stop** when credits reach zero
  • **Auto-shutdown timers** — set a max runtime when creating instances
  • **No debt** — we never charge more than your balance

    • The Result

    Our users report 40-60% cost savings compared to hourly-billed providers, especially during development and experimentation phases. The savings come from eliminating wasted minutes at the end of each session.

    Start with $5 in free credits and see the difference.

    Back to BlogGet Started Free