How much does an idle a4-highgpu-8g cost on Google Cloud?
An a4-highgpu-8g billed on-demand and left running around the clock costs about $94,082/mo at its cheapest listed region (us-central1). Scheduling it off outside business hours cuts that to roughly $28,001/mo — a $66,082/mo saving.
Cheapest on-demand
$128.8800/hr
us-central1
Running 24/7
$94,082/mo
730-hour month
Business-hours savings
$66,082/mo
up to 70% off the 24/7 bill (est.)
Price by region
| Region | Hourly (on-demand) | Spot (typical) | ≈ Monthly (24/7) |
|---|---|---|---|
| us-central1 Cheapest | $128.8800/hr | $39.6336/hr | ≈ $94,082/mo |
| us-east1 | $128.8800/hr | $39.6336/hr | ≈ $94,082/mo |
| asia-southeast1 | $154.6560/hr | $39.2168/hr | ≈ $112,899/mo |
What a schedule saves
| Schedule | Monthly cost | Savings vs 24/7 |
|---|---|---|
| Running 24/7 | $94,082/mo | baseline |
| Weekdays only (off on weekends) | $67,202/mo | -$26,881/mo (29%) |
| Business hours only (10×5) | $28,001/mo | -$66,082/mo (70%) |
Assumes a 730-hour month at on-demand list price. Actual savings depend on when the instance is actually used.
About the a4 family
A4 is GCP's B200 (Blackwell) machine, priced per GPU slice rather than as separate vCPU, RAM and GPU parts — eight slices make one a4-highgpu-8g. Much of the capacity is sold through reservations and scheduled (DWS) windows, which is why on-demand rates appear in so few regions. Blackwell kept warm between runs is the most expensive idle on the platform, and the bundled pricing means no part of it can be trimmed while the machine stays on.
Regional pricing varies: a4-highgpu-8g runs $128.8800/hr in us-central1, the cheapest region priced here, versus $154.6560/hr in asia-southeast1, the priciest — a 17% spread across regions for the identical instance type.
Common questions
How much does an idle a4-highgpu-8g cost per month?
At the cheapest listed region (us-central1), an a4-highgpu-8g costs $128.8800/hr, or about $94,082/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 $66,082/mo (70%) versus running 24/7. A lighter schedule — on around the clock on weekdays, off on weekends — still saves about $26,881/mo (29%).
Why is a4-highgpu-8g priced per GPU slice?
Google lists A4 as one SKU per B200 slice that already includes the machine's share of vCPUs and memory, so the instance price is simply eight slices; there are no separate core or RAM rates to compose. That also means nothing can be detached to save money — the only lever is whether the machine is running, which is why an expiry on every A4 lease matters more than on any smaller box.
Stop paying for idle Google Cloud 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 standard 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.