How much does an idle p5.48xlarge cost on AWS?
A p5.48xlarge billed on-demand and left running around the clock costs about $40,179/mo at its cheapest listed region (us-east-1). Scheduling it off outside business hours cuts that to roughly $11,958/mo — a $28,221/mo saving.
Cheapest on-demand
$55.0400/hr
us-east-1
Running 24/7
$40,179/mo
730-hour month
Business-hours savings
$28,221/mo
up to 70% off the 24/7 bill (est.)
Price by region
| Region | Hourly (on-demand) | Spot (typical) | ≈ Monthly (24/7) |
|---|---|---|---|
| us-east-1 Cheapest | $55.0400/hr | $23.6672/hr | ≈ $40,179/mo |
| us-east-2 | $55.0400/hr | $14.3104/hr | ≈ $40,179/mo |
| us-west-2 | $55.0400/hr | $17.0624/hr | ≈ $40,179/mo |
| ap-south-1 | $66.0480/hr | $33.6845/hr | ≈ $48,215/mo |
| ap-northeast-1 | $68.8000/hr | $19.2640/hr | ≈ $50,224/mo |
| ap-southeast-2 | $71.5520/hr | $32.9139/hr | ≈ $52,233/mo |
| sa-east-1 | $92.4672/hr | $81.3711/hr | ≈ $67,501/mo |
What a schedule saves
| Schedule | Monthly cost | Savings vs 24/7 |
|---|---|---|
| Running 24/7 | $40,179/mo | baseline |
| Weekdays only (off on weekends) | $28,699/mo | -$11,480/mo (29%) |
| Business hours only (10×5) | $11,958/mo | -$28,221/mo (70%) |
Assumes a 730-hour month at on-demand list price. Actual savings depend on when the instance is actually used.
About the p5 family
P5 packs eight NVIDIA H100 GPUs per instance — AWS's workhorse tier for serious model training, rented by the run rather than kept as a pet. The waste pattern is intent drift, not amnesia: the box is held "for the next run", the next run slips, and the meter never noticed the difference. Per idle hour it sits at the very top of most accounts.
Regional pricing varies: p5.48xlarge runs $55.0400/hr in us-east-1, the cheapest region priced here, versus $92.4672/hr in sa-east-1, the priciest — a 40% spread across regions for the identical instance type.
Common questions
How much does an idle p5.48xlarge cost per month?
At the cheapest listed region (us-east-1), a p5.48xlarge costs $55.0400/hr, or about $40,179/mo if it runs 24/7. That's the on-demand list price — an idle instance costs exactly the same as a busy one, since cloud billing doesn't know the difference.
How much can I save by scheduling it off outside work hours?
Restricting it to business hours only (10×5) saves about $28,221/mo (70%) versus running 24/7. A lighter schedule — on around the clock on weekdays, off on weekends — still saves about $11,480/mo (29%).
Do teams really forget p5.48xlarge instances running?
A lease with a hard stop turns drift into a deliberate choice: extending takes a click, while doing nothing means the instance stops itself and the meter stops with it. On an eight-GPU H100 box, that click-versus-nothing choice is worth more per hour than almost anywhere else in the fleet.
Stop paying for idle AWS servers
Idlefy runs a free audit of your fleet, then flips dev servers to stopped-by-default: lease one when you need it, and
when the lease ends it shuts itself off — stop, not terminate, persistent disks intact. Timers are set by your team, never
guessed, and Idlefy only manages instances tagged idlefy=enabled — on AWS that boundary is enforced by the IAM policy itself.
Prices are on-demand list price for Linux instances and exclude storage, network, data transfer, and any negotiated or volume discounts. Actual bills vary by usage and account-level pricing. Prices as of 2026-08-31. Spot figures are indicative — AWS values are derived from AWS's published typical savings vs on-demand for the region; GCP values are the listed Spot price. Spot capacity can be reclaimed by the provider. Stopping preserves EBS and persistent-disk volumes; data on local instance-store or local-SSD drives does not survive a stop.