CreateOS vs sfcompute

Sandbox Benchmark Comparison · Updated October 30, 2026

CreateOS and sfcompute were each faster in 1 of 2 shared sandbox benchmarks.

Head to head

  • Burst TTITime to interactive · lower is better
    CreateOS faster2.37x
    CreateOS
    136ms
    sfcompute
    322ms
  • DAXTotal workload time · lower is better
    sfcompute faster1.28x
    CreateOS
    53.21s
    sfcompute
    41.72s

Burst TTI

October 30, 2026

CreateOS

100% success
Median
136ms
p95
176ms
p99
178ms
CreateOS Burst TTI results

sfcompute

100% success
Median
322ms
p95
467ms
p99
494ms
sfcompute Burst TTI results

Median time to interactive over time

12 runs · September 4, 2026 to October 30, 2026
CreateOSsfcompute

CreateOS

100% success · October 30, 2026
Median
53.21s
CreateOS DAX results

sfcompute

100% success · October 9, 2026
Median
41.72s
sfcompute DAX results

Median total workload time over time

5 runs · September 16, 2026 to October 9, 2026
CreateOSsfcompute

Cost per DAX run

CreateOS

Per-second wall clock
$0.0098per run
53.21s median at $0.6624/hr

$0.0504/vCPU-hr + $0.0162/GiB-hr — parity with E2B. No egress fees.

Source pricing page

sfcompute

Per GiB-hour (parkable)
$0.0085per run
41.72s median at $0.7315/hr

Rollout sandboxes at $0.04572/GiB-hr while active (16 GiB); parked/idle drops to ~$0.000139/GiB-hr. Interactive cpu-8 node is $0.4824/hr with a 15-min minimum.

Source pricing page
Compare sandbox costs across all providers →
Work at CreateOS or sfcompute? See where you stand every day, with full run logs, not just on benchmark Friday.Get stealth benchmarks →

Methodology

  • One workload, every provider. The same script, payload, and model hit every provider, addressed the way its own API expects. Nobody gets a tuned or special path.
  • Repeated on a schedule. Every suite runs many iterations per provider, automated on a recurring schedule, so results reflect steady state rather than one lucky request.
  • Raw results published. Every run is committed as JSON to the public benchmarks repo, next to the code that produced it. Nothing is cherry-picked or held back.
  • How this comparison is built. Each benchmark compares the two providers' medians; results within 5% count as roughly even. A comparison is published only when both providers completed at least 50% of runs in 2 or more shared benchmarks run in the last 30 days. If a provider’s latest run of a benchmark failed, its most recent successful run from that window is used instead.

Frequently asked questions

Have another question? Email us.

CreateOS and sfcompute were each faster in 1 of 2 shared sandbox benchmarks. In Burst TTI (October 30, 2026), CreateOS had a median time to interactive of 136ms versus 322ms for sfcompute, 2.37x faster. In DAX (October 9, 2026), sfcompute had a median total workload time of 41.72s versus 53.21s for CreateOS, 1.28x faster.

PartnersGoogle Cloud Run