US
N. Virginia
us-east-1
spot · 7-day average $0.0137/hr
on-demand$0.0336
spot interrupt freq 15-20%
t4g.medium
AWS Graviton2 processor (64-bit Arm Neoverse cores). Burstable General Purpose instance type, utilizing a CPU credit model to provide a baseline level of performance with the ability to burst above it. Built on the AWS Nitro System. Offers up to 40% better price performance than T3 instances.
available for self-hosted runners on RunsOn
single-thread · passmark
1,105pts
vcpus
2
2 cores
memory
4 GiB
ram
clock speed
2.5 GHz
max frequency
architecture
arm64
Graviton
Measured
2 jobs of 2 finished in the EC2 CPU sweep, 30 Sep–10 Oct: the values are their medians. Every job links its log on the open harness.
CPU Mark
1,333
PassMark, both vCPUs
cores of work
1.93
sysbench, 2 threads ÷ 1
sysbench 1 thread
2,782
events per second
7-zip
8,561
MIPS, all threads
| date | suite | single-thread | capacity | job log |
|---|---|---|---|---|
| 10 Oct 2026 | EC2 CPU sweep | 1,108 | spot | job log ↗ |
| 30 Sep 2026 | EC2 CPU sweep | 1,103 | on-demand | job log ↗ |
- Burstable: a t type runs at full speed only while it has CPU credits. A benchmark this short sees the burst, not what an hour at this price sustains, so the board leaves it out of CPU per dollar.
EC2 instance family: t4g
What the family is built for, the silicon underneath, and how to ask RunsOn for it.
t4g burstable
cpu type
Neoverse-N1 — AWS Graviton2 processor (64-bit Arm Neoverse cores)
architecture
arm64
highlights
Burstable General Purpose instance type, utilizing a CPU credit model to provide a baseline level of performance with the ability to burst above it. Built on the AWS Nitro System. Offers up to 40% better price performance than T3 instances
released
September 2020
.github/workflows/ci.yml yaml
jobs: build: runs-on: - runs-on=${{ github.run_id }} - family=t4g - cpu=2
the label is a query — each line one constraint, resolved at launch. job labels →
Availability & price by region
Spot is the 7-day average of the cheapest zone's spot price in each region, from the RunsOn finder: an estimate of an hour in your own AWS account, where RunsOn adds no markup. The bar shows how far it runs under on-demand; the badge is AWS's spot-interruption band for the type.
US cheapest spot
Oregon
us-west-2
spot · 7-day average $0.0122/hr
on-demand$0.0336
spot interrupt freq 15-20%
IE
Ireland
eu-west-1
spot · 7-day average $0.0168/hr
on-demand$0.0368
spot interrupt freq >20%
DE
Frankfurt
eu-central-1
spot · 7-day average $0.0172/hr
on-demand$0.0384
spot interrupt freq >20%
Spot is the cheapest availability zone's spot price at each of the RunsOn finder's fetches (every 90 minutes), averaged over 7 days; on-demand is the Linux list price. Savings compare the two. Prices fetched 12 Oct 2026. The interruption band is AWS Spot Advisor's, fetched 12 Oct 2026.
About these benchmarks
The open benchmark harness launches t4g.medium as a RunsOn runner and runs PassMark PerformanceTest (CPU Single
Threaded and CPU Mark), sysbench on one thread and on every thread, and 7-Zip's built-in benchmark. These numbers come from the
EC2 CPU sweep, the suite that measured this type in the most runs; each is the median over its runs of the current
workload definition in the last 75 days.
A type takes a place once it has 3 runs. A type within 5% of the type heading its group shares that place (“joint #2”); a lead under 5% is no lead. Burstable types are listed and ranked on CPU, never on CPU per dollar.
Prices come from the RunsOn finder for the four regions shown; savings compare its 7-day average spot price with the on-demand price. RunsOn launches this type the same way for any job: ephemeral EC2 in your account, one CloudFormation stack. More in the methodology.
Run t4g.medium in your own account.
The benchmark runs on the same RunsOn anyone can deploy. One CloudFormation stack, your AWS account, ~10 minutes.