How much does an idle p6-b200.48xlarge cost on AWS?
A p6-b200.48xlarge billed on-demand and left running around the clock costs about $83,171/mo at its cheapest listed region (us-east-1). Scheduling it off outside business hours cuts that to roughly $24,753/mo — a $58,418/mo saving.
Cheapest on-demand
$113.9328/hr
us-east-1
Running 24/7
$83,171/mo
730-hour month
Business-hours savings
$58,418/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 | $113.9328/hr | $23.9259/hr | ≈ $83,171/mo |
| us-east-2 | $113.9328/hr | $20.5079/hr | ≈ $83,171/mo |
| us-west-2 | $113.9328/hr | $20.5079/hr | ≈ $83,171/mo |
| ap-south-1 | $160.6453/hr | — | ≈ $117,271/mo |
What a schedule saves
| Schedule | Monthly cost | Savings vs 24/7 |
|---|---|---|
| Running 24/7 | $83,171/mo | baseline |
| Weekdays only (off on weekends) | $59,408/mo | -$23,763/mo (29%) |
| Business hours only (10×5) | $24,753/mo | -$58,418/mo (70%) |
Assumes a 730-hour month at on-demand list price. Actual savings depend on when the instance is actually used.
About the p6-b200 family
P6-B200 carries eight NVIDIA B200 GPUs — the Blackwell generation — at the very top of AWS's training tier. Capacity is scarce enough that teams who win quota hesitate to release an instance once they have it, and the "keep it warm for the next run" habit is at its most expensive here. If any machine in an account deserves a hard, automatic off-switch, it is this one.
Regional pricing varies: p6-b200.48xlarge runs $113.9328/hr in us-east-1, the cheapest region priced here, versus $160.6453/hr in ap-south-1, the priciest — a 29% spread across regions for the identical instance type.
Common questions
How much does an idle p6-b200.48xlarge cost per month?
At the cheapest listed region (us-east-1), a p6-b200.48xlarge costs $113.9328/hr, or about $83,171/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 $58,418/mo (70%) versus running 24/7. A lighter schedule — on around the clock on weekdays, off on weekends — still saves about $23,763/mo (29%).
Is Blackwell worth it over H100 or H200 for training?
Per unit of training throughput, newer silicon generally wins when you can actually get it — the real constraints are quota and regional availability, not benchmarks. The unromantic advice: benchmark your own workload, and whichever generation you land on, make sure the instance stops itself when the run ends. Idle Blackwell is the most expensive idle on the market.
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.