How much does an idle g6.12xlarge cost on AWS?
A g6.12xlarge billed on-demand and left running around the clock costs about $3,359/mo at its cheapest listed region (us-east-1). Scheduling it off outside business hours cuts that to roughly $1,000/mo — a $2,359/mo saving.
Cheapest on-demand
$4.6016/hr
us-east-1
Running 24/7
$3,359/mo
730-hour month
Business-hours savings
$2,359/mo
70% off the 24/7 bill
Price by region
| Region | Hourly (on-demand) | Spot (typical) | ≈ Monthly (24/7) |
|---|---|---|---|
| us-east-1 Cheapest | $4.6016/hr | $2.2548/hr | ≈ $3,359/mo |
| us-east-2 | $4.6016/hr | $1.3805/hr | ≈ $3,359/mo |
| us-west-2 | $4.6016/hr | $2.3468/hr | ≈ $3,359/mo |
| ap-south-1 | $5.5256/hr | $2.1550/hr | ≈ $4,034/mo |
| eu-central-1 | $5.7543/hr | $1.9565/hr | ≈ $4,201/mo |
| ap-southeast-2 | $5.9830/hr | $2.2735/hr | ≈ $4,368/mo |
| ap-northeast-1 | $6.6737/hr | $2.4693/hr | ≈ $4,872/mo |
| sa-east-1 | $7.8218/hr | $1.3297/hr | ≈ $5,710/mo |
What a schedule saves
| Schedule | Monthly cost | Savings vs 24/7 |
|---|---|---|
| Running 24/7 | $3,359/mo | baseline |
| Weekdays only (off on weekends) | $2,399/mo | -$960/mo (29%) |
| Business hours only (10×5) | $1,000/mo | -$2,359/mo (70%) |
Assumes a 730-hour month at on-demand list price. Actual savings depend on when the instance is actually used.
About the g6 family
G6 uses NVIDIA's current-generation L4 GPUs for the best price-performance in the inference and graphics tier, increasingly the default for new ML-serving and rendering environments. Being newest, these instances are the least likely to have inherited a shutdown schedule, so fresh g6 dev and staging boxes routinely start life running 24/7.
Regional pricing varies: g6.12xlarge runs $4.6016/hr in us-east-1, the cheapest region priced here, versus $7.8218/hr in sa-east-1, the priciest — a 41% spread across regions for the identical instance type.
The next size up in the same family, g6.16xlarge (64 vCPU / 256 GB RAM), costs about $2,480/mo running 24/7 at its cheapest region — roughly 26% less than g6.12xlarge's ~$3,359/mo — for 1.3x the vCPU. Note the trade: it carries 1 GPU versus 4 on the g6.12xlarge — cheaper only if your work isn't GPU-bound.
Common questions
How much does an idle g6.12xlarge cost per month?
At the cheapest listed region (us-east-1), a g6.12xlarge costs $4.6016/hr, or about $3,359/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 $2,359/mo (70%) versus running 24/7. A lighter schedule — on around the clock on weekdays, off on weekends — still saves about $960/mo (29%).
Is g6 cheaper to run than g4dn for inference?
Per unit of inference work, usually yes — the L4 delivers considerably more throughput per dollar than the T4 at a moderately higher hourly rate. For always-on production serving that efficiency compounds; for a dev box used a few hours a day the hourly rate dominates, and a stopped g4dn is still cheaper than a running g6.
Other g6 sizes
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 2026-07-20. 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.