Speed
1.7×
faster than GitHub-hosted (ubuntu-latest)
2nd of 9 by speed
Cost per job
−84%
vs GitHub-hosted (ubuntu-latest)
2nd of 9; the same at 1,000 to 100,000 jobs a month
Lowest median time
1 / 12
workloads
codex-lint
Failed runs
0 of 360
no failed measured run
across all workloads
Speed on each workload
× faster than GitHub-hosted (ubuntu-latest) on the same workloadConfiguration
Managed · up to 2 vCPU / 8 GB- Measured configuration
- Vitko Runners
- Size
- up to 2 vCPU / 8 GB
- Label or setup
Vitko Runners label- Price as measured
- $0.002 a minute, billed per second; list price on 2026-10-05 (source)
- Notes
- Every workload, codex-lint included, runs at 2 vCPU / 8 GiB; the guest has swap, as GitHub-hosted runners do (the probe records memory and swap per job).
- Definition
- providers/vitko.json
Per workload
Green marks the lowest median time on the workload. The burst row covers all 20 jobs.| Workload | Median time | vs GitHub-hosted (ubuntu-latest) | p95 time | Mean cost | vs GitHub-hosted (ubuntu-latest) | Successful runs | Rank |
|---|---|---|---|---|---|---|---|
| Burst: 20 jobs at once | 160 s | 1.6x faster | 372 s | $0.056 | 89% lower | 30 of 30 | 5th |
| cli | 272 s | 1.4x faster | 326 s | $0.0097 | 75% lower | 30 of 30 | 3rd |
| codex-lint | 404 s | 2.2x faster | 441 s | $0.014 | 84% lower | 30 of 30 | 1st |
| container | 179 s | 1.3x faster | 199 s | $0.0060 | 78% lower | 30 of 30 | 4th |
| fd | 49 s | 1.9x faster | 56 s | $0.0016 | 87% lower | 30 of 30 | 2nd |
| flask | 16 s | 1.4x faster | 20 s | $0.0005 | 92% lower | 30 of 30 | 3rd |
| guava | 655 s | 1.8x faster | 661 s | $0.022 | 80% lower | 30 of 30 | 2nd |
| monorepo | 180 s | 2.1x faster | 211 s | $0.0058 | 85% lower | 30 of 30 | 3rd |
| pydantic | 105 s | 1.7x faster | 140 s | $0.0034 | 83% lower | 30 of 30 | 3rd |
| ts | 372 s | 2.1x faster | 404 s | $0.012 | 84% lower | 30 of 30 | 2nd |
| vite | 350 s | 1.7x faster | 389 s | $0.012 | 81% lower | 30 of 30 | 3rd |
| vue | 63 s | 1.9x faster | 64 s | $0.0020 | 85% lower | 30 of 30 | 2nd |
- Burst queue, p95
- 295 s, against 31 s for GitHub-hosted (ubuntu-latest)
Indexes and 95% intervals
- Time index
- 0.574 (95% interval 0.556 to 0.602)
- Cost index
- 0.157 (0.155 to 0.159) at 100,000 jobs a month