Render Pricing Explained: How Prorated Billing Actually Works

Scaling your Render service up to a 2 GB instance shows a tidy $25/mo price tag, but that number is just a monthly equivalent. Render bills compute per second based on the actual time your instance is running, so part-time upgrades cost far less than the headline figure. Here’s the quick guide you can reference before bumping your plan for a burst of traffic or a load test.

How Render calculates compute costs

• Instance price is “full-month equivalent.” The $25 represents running the 2 GB RAM plan 24/7 for ~30 days.
• Prorated billing. Render tracks runtime in seconds. Spin up the plan for a few hours, then scale down, and you only pay for those hours.
• Separate line items. Databases, bandwidth overages, build minutes, and add-on services are metered independently. Compute time is just one part of your statement.

Real-world example

Say you upgrade a web service to the 2 GB plan for a ten-day campaign:

Runtime fraction = 10 days ÷ 30-day month ≈ 0.33
Prorated charge  = 0.33 × $25 ≈ $8.33

The invoice will show roughly eight bucks for compute rather than the full $25 plus whatever else you consumed that month (bandwidth, cron jobs, etc.). If you only needed the extra headroom for a weekend, the cost drops even further.

When per-second billing helps most

• Seasonal spikes. Scale up during product launches, waitlists, or marketing pushes, then revert to smaller instances when traffic normalizes.
• Load testing. Temporarily double resources without committing to a monthly upgrade.
• Development & previews. Keep staging environments cheap by running large instances only while you’re actively testing.

Quick reference for API & dashboard

To keep invoices tidy:

• Pause services when idle. Stopped services no longer accrue compute charges.
• Monitor in the dashboard. Team ➜ Billing shows accrued usage in real time.
• Watch other meters. Outbound data beyond the free allowance and extra build minutes are charged separately.

Wrap-up

Render’s pricing looks flat, but it’s really usage-based. Treat the advertised monthly price as the ceiling for 24/7 uptime. If you only need more horsepower for a slice of the month or a few hours on demand you’ll pay exactly for that slice. One more reason Render works well for teams that scale resources up and down frequently.

👉 Check the latest rates under Web Services on the Render pricing page.