Fastest GitHub Actions Runners in 2026

· RunsOn

10 min read Original article ↗

Single-thread CPU, queue time, and real $/min for every Linux runner worth considering — official GitHub, self-hosted, and third-party SaaS. The leaderboard re-ranks live as you drag the speed-vs-price dial, and RunsOn rows use real EC2 spot prices in your own AWS account.

TL;DR#

Best value

RunsOn, Namespace, and Blacksmith cluster in the cheap-and-fast corner. RunsOn leads self-hosted on both speed and price; AWS CodeBuild is slower and pricier.

Fastest CPU

x64: Namespace leads; Warpbuild on Ryzen 9950X now ranks second, narrowly ahead of Avrea, Blacksmith, and RunsOn m8azn. arm64: Avrea’s Apple M5 Max leads Namespace’s Apple M4, followed by RunsOn’s Neoverse V3.

Cheapest

Self-hosted spot on AWS lands ~5.9× under GitHub — the bill goes to your own AWS account, not a middleman.

Hardware caveat: some third-party SaaS runners offer their newest CPUs only when capacity is available, so peak-time jobs may land on slower hardware. RunsOn launches the EC2 family you select, so the instance type is known.

How to read: leaderboard and deep dive use the configured CPU/queue percentiles (p50/p50). Higher CPU is better, lower queue is better. Price is normalized $/min.

Providers included#

CPU speed vs price per minute#

Each dot is a runner. Up is faster, left is cheaper — so the top-left corner is where you want to be. Dot size is queue time; hover any dot or row to link the two. Drag the speed-vs-price dial and the leaderboard re-ranks live.

Last updated:

Benchmarks use Linux runners only, the Passmark single-thread metric, and the last 30 days of data.

cheaperfasterleaning faster
70% speed / 30% price

cpu × price × queuespot · us-east-1 · x64

★ cheap & fast15002000250030003500400045005000$0.0000$0.0013$0.0026$0.0039$0.0052$0.0065price / min (normalized to 2 vCPU / 8GB) → cheaper is leftCPU score → faster is upRunsOnGitHub

who's who

RunsOn — self-hosted, your AWS account

GitHub — official, the reference price

AWS CodeBuild — managed, in your account

SaaS — Namespace, Blacksmith, Avrea, Warpbuild, Ubicloud

dot size = queue time

Smaller is better — the wait between trigger and first step. GitHub queues fastest; self-hosted spot trades a few seconds for much lower cost.

rank by

#providertypespecscpuqueueprice / minvalue
1Namespacenscloud-ubuntu-24.04-amd64-2x8SaaSstandard4×8GBnorm $0.0024468610s$0.00401.5× cheaper90
2RunsOn family=m8aznSelf-hostedspot2×8GB426622s$0.00125.1× cheaper86
3Warpbuildwarp-ubuntu-2404-x64-2xSaaS2×7GB443314s$0.00401.5× cheaper78
4Avreaavrea-ubuntu-24.04-2-vcpuSaaS2×8GB435512s$0.00401.5× cheaper77
5Blacksmithblacksmith-2vcpu-ubuntu-2404SaaS2×8GB433014s$0.00401.5× cheaper76
6RunsOn family=m8aSelf-hostedspot2×8GB380322s$0.00096.9× cheaper74
7Ubicloudubicloud-premium-2-ubuntu-2404SaaSpremium2×8GB352923s$0.00203.0× cheaper60
8Warpbuildwarp-ubuntu-2404-x64-2xSaaS2×7GB378810s$0.00401.5× cheaper60
9RunsOn family=m8i+r8iSelf-hostedspot2×8GB319524s$0.00105.9× cheaper56
10StarSlingstarsling-ubuntu-24.04SaaS2×8GB335512s$0.00401.5× cheaper47
11Ubicloudubicloud-standard-2-ubuntu-2404SaaSstandard2×8GB292324s$0.00134.8× cheaper47
12RunsOn family=c7a+m7a+r7aSelf-hostedspot2×8GB287426s$0.00106.0× cheaper47
13Ubicloudubicloud-standard-2-ubuntu-2404SaaSstandard2×8GB290732s$0.00134.8× cheaper45
14RunsOn family=c7i+m7i+r7iSelf-hostedspot2×8GB271425s$0.00096.5× cheaper43
15GitHububuntu-24.04, github24-2cpu-x64Official2×8GB35006s$0.0060≈ GitHub42
16RunsOn family=c6a+m6a+r6aSelf-hostedspot2×8GB258025s$0.00106.2× cheaper39
17GitHubgithub24-2cpu-x64, ubuntu-24.04Official2×8GB33587s$0.0060≈ GitHub37
18GitHubgithub24-2cpu-x64, ubuntu-24.04Official2×8GB23517s$0.0060≈ GitHub6
19GitHububuntu-24.04, github24-2cpu-x64Official2×8GB22696s$0.0060≈ GitHub4
20GitHubgithub24-2cpu-x64, ubuntu-24.04Official2×8GB21976s$0.0060≈ GitHub2
21AWS CodeBuildcodebuild-ubuntu-8.0-large, codebuild-ubuntu-8.0-small, codebuild-ubuntu-8.0-mediumSelf-hosted2×4GBnorm $0.0059214124s$0.00501.2× cheaper1
22AWS CodeBuildcodebuild-ubuntu-8.0-medium, codebuild-ubuntu-8.0-large, codebuild-ubuntu-8.0-smallSelf-hosted2×4GBnorm $0.0059201924s$0.00501.2× cheaper1

How to choose#

$/minWant the lowest bill? Sort by cheapest. Self-hosted spot on AWS lands ~5.9× under GitHub — the cost goes to your AWS account, not a middleman.

cpuWant raw speed? Sort by fastest CPU. Namespace leads x64, followed by Warpbuild on Ryzen 9950X, Avrea, Blacksmith, and RunsOn m8azn.

queueLatency-sensitive? Sort by lowest queue. GitHub starts fastest; for self-hosted, warm pools cut RunsOn queues to under ~10s.

ctrlWant control & isolation? Pick self-hosted. RunsOn keeps runners, logs, caches and secrets inside your own VPC — no third party touches your code. See networking options.

Deep dive (full data)#

Open full benchmark table with distributions

Includes per-runner distributions, CPU/queue percentiles, and provider metadata for the last 30 days.

# Provider CPU speed (p50) Queue time (p50) + 30d P90 trend Processor
1

Namespace SaaS

nscloud-ubuntu-24.04-amd64-2x8

$0.004/min 1.5x cheaper standard

p50: 4686p90: 4717

p50: 10.0sp90: 13.0s

AMD EPYC (x86_64)

US/Europe (might vary)

29 samples


CPU 4500+


2

Warpbuild SaaS

warp-ubuntu-2404-x64-2x

$0.004/min 1.5x cheaper

p50: 4284p90: 4515

p50: 14.0sp90: 16.0s

AMD Ryzen 9 9950X 16-Core Processor(x86_64)

Hetzner / AWS / BYOC

20 samples

3

Avrea SaaS

avrea-ubuntu-24.04-2-vcpu

$0.004/min 1.5x cheaper

p50: 4354p90: 4484

p50: 12.0sp90: 15.0s

AMD EPYC 4585PX 16-Core Processor(x86_64)

Dedicated hardware

28 samples

4

Blacksmith SaaS

blacksmith-2vcpu-ubuntu-2404

$0.004/min 1.5x cheaper

p50: 4282p90: 4482

p50: 14.0sp90: 18.0s

AMD EPYC (x86_64)

Hetzner

28 samples

5

RunsOn Self-Hosted

runs-on,family=m8azn

$0.0012/min 5.1x cheaper spot

p50: 4266p90: 4279

p50: 22.0sp90: 29.0s

AMD EPYC 9R05 (x86_64)

AWS

29 samples


CPU 4000+


6

RunsOn Self-Hosted

runs-on,family=m8a

$0.0009/min 6.9x cheaper spot

p50: 3803p90: 3827

p50: 22.0sp90: 40.0s

AMD EPYC 9R45 (x86_64)

AWS

29 samples

7

Warpbuild SaaS

warp-ubuntu-2404-x64-2x

$0.004/min 1.5x cheaper

p50: 3788p90: 3873

p50: 10.0sp90: 18.0s

AMD Ryzen 9 7950X3D 16-Core Processor(x86_64)

Hetzner / AWS / BYOC

9 samples

8

Ubicloud SaaS

ubicloud-premium-2-ubuntu-2404

$0.002/min 3.0x cheaper premium

p50: 3529p90: 3738

p50: 23.0sp90: 34.0s

AMD Ryzen 9 7950X3D 16-Core Processor(x86_64)

Hetzner (Germany)

29 samples

9

GitHub Official

ubuntu-24.04, github24-2cpu-x64

$0.006/min ≈ GitHub

p50: 3500p90: 3596

p50: 6.0sp90: 8.0s

AMD EPYC 9V45 96-Core Processor(x86_64)

Azure

3 samples


CPU 3500+


10

GitHub Official

github24-2cpu-x64, ubuntu-24.04

$0.006/min ≈ GitHub

p50: 3358p90: 3382

p50: 7.0sp90: 7.0s

Intel(R) Xeon(R) 6973P-C (x86_64)

Azure

3 samples

11

StarSling SaaS

starsling-ubuntu-24.04

$0.004/min 1.5x cheaper

p50: 3355p90: 4089

p50: 12.0sp90: 14.0s

AMD EPYC (x86_64)

Oracle Cloud

29 samples

12

RunsOn Self-Hosted

runs-on,family=m8i+r8i

$0.001/min 5.9x cheaper spot

p50: 3195p90: 3199

p50: 24.0sp90: 43.0s

Intel(R) Xeon(R) 6975P-C (x86_64)

AWS

29 samples


CPU 3000+


13

Ubicloud SaaS

ubicloud-standard-2-ubuntu-2404

$0.0013/min 4.8x cheaper standard

p50: 2918p90: 2945

p50: 24.0sp90: 34.0s

AMD EPYC 9454P 48-Core Processor(x86_64)

Hetzner (Germany)

24 samples

14

Ubicloud SaaS

ubicloud-standard-2-ubuntu-2404

$0.0013/min 4.8x cheaper standard

p50: 2907p90: 2937

p50: 32.0sp90: 40.0s

AMD EPYC 9454 48-Core Processor(x86_64)

Hetzner (Germany)

3 samples

15

RunsOn Self-Hosted

runs-on,family=c7a+m7a+r7a

$0.001/min 6.0x cheaper spot

p50: 2874p90: 2878

p50: 26.0sp90: 51.0s

AMD EPYC 9R14 (x86_64)

AWS

29 samples

16

RunsOn Self-Hosted

runs-on,family=c7i+m7i+r7i

$0.0009/min 6.5x cheaper spot

p50: 2714p90: 2919

p50: 25.0sp90: 30.0s

Intel(R) Xeon(R) Platinum 8488C (x86_64)

AWS

29 samples

17

RunsOn Self-Hosted

runs-on,family=m6a

$0.001/min 6.2x cheaper spot

p50: 2580p90: 2586

p50: 25.0sp90: 26.0s

AMD EPYC 7R13 Processor (x86_64)

AWS

29 samples


CPU 2500+


18

GitHub Official

github24-2cpu-x64, ubuntu-24.04

$0.006/min ≈ GitHub

p50: 2351p90: 2828

p50: 7.0sp90: 7.0s

INTEL(R) XEON(R) PLATINUM 8573C (x86_64)

Azure

3 samples

19

GitHub Official

ubuntu-24.04, github24-2cpu-x64

$0.006/min ≈ GitHub

p50: 2269p90: 2277

p50: 6.0sp90: 8.0s

AMD EPYC 7763 64-Core Processor(x86_64)

Azure

33 samples

20

GitHub Official

github24-2cpu-x64, ubuntu-24.04

$0.006/min ≈ GitHub

p50: 2197p90: 2788

p50: 6.0sp90: 8.0s

AMD EPYC 9V74 80-Core Processor(x86_64)

Azure

15 samples

21

AWS CodeBuild Self-Hosted

codebuild-ubuntu-8.0-large, codebuild-ubuntu-8.0-small, codebuild-ubuntu-8.0-medium

$0.005/min 1.2x cheaper

p50: 2140p90: 2154

p50: 24.0sp90: 29.0s

Intel(R) Xeon(R) Platinum 8275CL CPU @ 3.00GHz (x86_64)

AWS

40 samples

22

AWS CodeBuild Self-Hosted

codebuild-ubuntu-8.0-medium, codebuild-ubuntu-8.0-large, codebuild-ubuntu-8.0-small

$0.005/min 1.2x cheaper

p50: 2019p90: 2039

p50: 24.0sp90: 27.0s

Intel(R) Xeon(R) Platinum 8124M CPU @ 3.00GHz (x86_64)

AWS

47 samples

# Provider CPU speed (p50) Queue time (p50) + 30d P90 trend Processor
1

Avrea SaaS

avrea-ubuntu-24.04-arm-2-vcpu

$0.012/min 2.4x pricier

p50: 4840p90: 4897

p50: 18.0sp90: 20.0s

Apple M5 Max + Virtualization

Dedicated hardware

30 samples

2

Namespace SaaS

nscloud-ubuntu-24.04-arm64-2x8,namespace-features:linux-on-apple-silicon=true

$0.012/min 2.4x pricier premium

p50: 4802p90: 4924

p50: 12.0sp90: 24.0s

Apple M4 + Virtualization

US/Europe (might vary)

29 samples


CPU 4500+


3

RunsOn Self-Hosted

runs-on,family=m9g

$0.0009/min 5.5x cheaper spot

p50: 2462p90: 2467

p50: 24.0sp90: 45.0s

Neoverse-V3 (aarch64)

AWS

29 samples


CPU 2000+


4

Warpbuild SaaS

warp-ubuntu-latest-arm64-2x

$0.003/min 1.7x cheaper

p50: 1944p90: 1945

p50: 19.0sp90: 40.0s

Neoverse-V2 (aarch64)

Hetzner / AWS / BYOC

30 samples

5

RunsOn Self-Hosted

runs-on,family=c8g+m8g+r8g

$0.0009/min 5.7x cheaper spot

p50: 1945p90: 1947

p50: 24.0sp90: 34.0s

Neoverse-V2 (aarch64)

AWS

58 samples

6

GitHub Official

github24-2cpu-arm64

$0.005/min ≈ GitHub

p50: 1874p90: 1878

p50: 8.0sp90: 9.0s

Neoverse-N2 (aarch64)

Azure

30 samples

7

RunsOn Self-Hosted

runs-on,family=c7g+m7g+r7g

$0.001/min 5.1x cheaper spot

p50: 1557p90: 1561

p50: 25.0sp90: 45.0s

Neoverse-V1 (aarch64)

AWS

29 samples


CPU 1500+


8

Blacksmith SaaS

blacksmith-2vcpu-ubuntu-2404-arm

$0.0025/min 2.0x cheaper

p50: 1347p90: 1375

p50: 21.0sp90: 23.0s

Ampere-1a (aarch64)

Hetzner

26 samples

9

Blacksmith SaaS

blacksmith-2vcpu-ubuntu-2404-arm

$0.0025/min 2.0x cheaper

p50: 1325p90: 1332

p50: 18.0sp90: 19.0s

Neoverse-N1 (aarch64)

Hetzner

4 samples

10

Ubicloud SaaS

ubicloud-standard-2-arm-ubuntu-2404

$0.0013/min 4.0x cheaper standard

p50: 1309p90: 1315

p50: 13.0sp90: 34.0s

Neoverse-N1 (aarch64)

Hetzner (Germany)

30 samples

11

Namespace SaaS

nscloud-ubuntu-24.04-arm64-2x8

$0.004/min 1.3x cheaper standard

p50: 1229p90: 1261

p50: 20.0sp90: 26.0s

Ampere-1a (aarch64)

US/Europe (might vary)

16 samples

What the data says#

Performance

  • Namespace leads x64; Warpbuild on Ryzen 9950X ranks second, followed by Avrea, Blacksmith, and RunsOn m8azn.
  • Avrea’s Apple M5 Max leads arm64, followed by Namespace’s Apple M4 and RunsOn’s Neoverse V3.
  • GitHub runs older CPUs overall, most visibly on x64.

Reliability & scaling

  • GitHub’s queue times are excellent for standard runners — and much improved for larger ones.
  • AWS scales better than Hetzner-backed fleets under load.
  • Some SaaS providers gate higher vCPU concurrency (and static egress IPs) behind requests or fees; self-hosted RunsOn and CodeBuild aren’t affected.
  • CPU models on third-party SaaS runners can vary with capacity: the newest hardware may be available off-peak, while peak-time jobs land on slower CPUs. RunsOn launches the EC2 family you select, so the instance type is explicit.

Security & infra

  • RunsOn is fully self-hosted in your AWS account — your IAM, your network, a dedicated GitHub App.
  • AWS CodeBuild is managed by AWS but runs in your account (shared GitHub App for registration).
  • GitHub runners are managed by GitHub on Azure; private access needs secrets or OIDC.
  • Namespace, Blacksmith, Warpbuild, Ubicloud and Depot are SaaS; control plane and registration aren’t fully yours.

Gap: detailed concurrency/scaling tests are still missing; they matter for high-volume CI/CD (thousands of jobs/day). RunsOn routinely handles more than 1M jobs every day.

Benchmark changelog#

Routine benchmark sample refreshes, design-only changes, SEO edits, and internal refactors are omitted. These are the public-facing changes that affected provider coverage, pricing assumptions, scoring, or data filtering.

  • Aug 2024 — The benchmark tracked CPU speed and queue time across the initial runner-provider set.
  • Oct 2024 — Results moved toward p50/p95 CPU and queue distributions, and restricted or unavailable provider rows were handled explicitly.
  • Jul 2025 — Provider summaries were revamped, and the median calculation was corrected.
  • Oct 2025 — RunsOn m8a family results were added.
  • Feb 2026 — The pricing-aware leaderboard was added, including provider pricing metadata, CPU-vs-price ranking, RunsOn EBS/license costs, and Cirrus monthly pricing normalization.
  • Feb 2026 — RunsOn family-specific pricing, m8azn, spot/on-demand handling, p50/p90 CPU and queue percentiles, queue penalties, value scoring fixes, and family exclusions were added.
  • Apr 2026 — BuildJet and Cirrus were removed from active benchmark/pricing coverage; Cirrus was removed after Cirrus Labs announced it was joining OpenAI and Cirrus Runners stopped accepting new customers.
  • May 2026 — Shipfox was removed. Ubicloud standard runner pricing was updated from $0.0008 to $0.0010 per minute for the 2 vCPU runner after its 2026 price adjustment; premium runner pricing was unchanged.
  • Jun 2026 — m9g was added, and value scoring started normalizing price by runner resources so differently sized runner shapes compare more fairly.
  • Jun 2026 — Avrea was added with published pay-as-you-go pricing and Apple M5 Max ARM labeling.
  • Jul 2026 — StarSling was added with published pay-as-you-go pricing for x64 Linux runners.
  • Aug 2026 — Ubicloud standard and premium runner prices were increased by 25% ahead of the September 1 effective date. The 2 vCPU rates moved from $0.0010 to $0.00125 per minute for standard runners and from $0.0016 to $0.0020 per minute for premium runners.

Fastest self-hosted, at ~5.9× lower cost.

The benchmark runs on the same RunsOn anyone can deploy. One CloudFormation stack, your account, ~10 minutes.

Note: this analysis is updated regularly. Things change quickly in this space, so make sure you do your own research as well.