Cloud instance idle costs

An idle server costs exactly the same as a busy one — cloud billing doesn't know the difference. We priced 117 AWS and Google Cloud instance types at their cheapest listed region, both running 24/7 and scheduled off outside work hours, so you can see the real monthly cost — and the savings — before you provision anything.

Commonly looked-up instance types

Monthly cost at each type's cheapest listed region, running non-stop versus running business hours only (50 hours a week). Every type below has its own page with a full region breakdown, plus typical spot pricing where the provider publishes it.

Instance type Specs 24/7 Business hours Saved / mo
t3.medium AWS 2 vCPU · 4 GB $30 $9 $21
t3.large AWS 2 vCPU · 8 GB $61 $18 $43
m5.large AWS 2 vCPU · 8 GB $70 $21 $49
m5.xlarge AWS 4 vCPU · 16 GB $140 $42 $98
m5.2xlarge AWS 8 vCPU · 32 GB $280 $83 $197
c5.xlarge AWS 4 vCPU · 8 GB $124 $37 $87
c6i.2xlarge AWS 8 vCPU · 16 GB $248 $74 $174
r5.xlarge AWS 4 vCPU · 32 GB $184 $55 $129
m7i.large AWS 2 vCPU · 8 GB $74 $22 $52
g4dn.xlarge AWS 4 vCPU · 16 GB · 1 GPU $384 $114 $270
g5.xlarge AWS 4 vCPU · 16 GB · 1 GPU $734 $219 $516
p3.2xlarge AWS 8 vCPU · 61 GB · 1 GPU $2,234 $665 $1,569
e2-standard-2 GCP 2 vCPU · 8 GB $49 $15 $34
e2-standard-4 GCP 4 vCPU · 16 GB $98 $29 $69
e2-standard-8 GCP 8 vCPU · 32 GB $196 $58 $137
n2-standard-4 GCP 4 vCPU · 16 GB $142 $42 $100
n2-standard-8 GCP 8 vCPU · 32 GB $284 $84 $199
c2-standard-8 GCP 8 vCPU · 32 GB $305 $91 $214
n1-standard-4 GCP 4 vCPU · 15 GB $139 $41 $97
e2-highmem-4 GCP 4 vCPU · 32 GB $132 $39 $93

Not the type you were looking for? Browse all AWS instance costs or Google Cloud instance costs.

How these numbers are calculated

Every figure starts from the provider's published on-demand hourly rate for Linux instances — no reserved-instance, savings-plan, or committed-use discount is applied. A month is 730 hours, the same convention AWS and Google Cloud use on their own pricing pages, so a 24/7 monthly cost is simply the hourly rate times 730.

The comparison column assumes the machine runs only during business hours — 50 of the 168 hours in a week. That is a 70% reduction in running hours, and because Linux compute is billed per second while a machine is running, the cost falls by the same proportion. Stopping a machine does not stop its disk charges, which is why storage sits outside these numbers.

Each instance page quotes the cheapest region we track for that type, alongside a full per-region table and typical spot pricing where the provider publishes it. Regions differ by a wide margin — the same type can cost noticeably more in São Paulo or Tokyo than in Northern Virginia — so compare within the region you actually deploy to.

Calculate your own savings

Prefer to work from your own fleet numbers instead of a single instance type? Use one of the calculators below.

Stop paying for idle 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.

Run a free idle audit

New to the lease model? See how Idlefy works →

Prices are on-demand list price and exclude storage, network, data transfer, and any negotiated or volume discounts. Actual bills vary by usage and account-level pricing. Stopping preserves EBS and persistent-disk volumes; data on local instance-store or local-SSD drives — which several of the GPU types above carry — does not survive a stop. Prices as of 2026-07-20.