Sandbox Benchmarks
A leaderboard of common benchmarks for each of our sandbox providers.
Provider Leaderboard
Performance Over Time
Composite Score
Detailed Metrics
Provider | Score | Median | P95 | P99 | Success |
|---|---|---|---|---|---|
| Miosa | 98.8 | 0.12s | 0.13s | 0.13s | 100% |
| Isorun | 98.6 | 0.14s | 0.15s | 0.15s | 100% |
| CreateOS | 98.3 | 0.16s | 0.18s | 0.18s | 100% |
| Archil | 97.2 | 0.19s | 0.39s | 0.44s | 100% |
| GravixlayerPre-release | 96.7 | 0.33s | 0.34s | 0.34s | 100% |
| Arker | 96.4 | 0.35s | 0.37s | 0.37s | 100% |
| sfcompute | 94.9 | 0.43s | 0.60s | 0.63s | 100% |
| Superserve | 94.2 | 0.52s | 0.64s | 0.73s | 100% |
| Cloud Run | 92.6 | 0.62s | 0.91s | 0.96s | 100% |
| Cloudflare | 92.2 | 0.64s | 0.73s | 0.84s | 99% |
| Daytona | 92.7 | 0.65s | 0.85s | 0.87s | 100% |
| Vercel | 92.6 | 0.67s | 0.82s | 0.91s | 100% |
| Tenki | 91.1 | 0.86s | 0.92s | 0.95s | 100% |
| Beam | 90.2 | 0.87s | 1.13s | 1.14s | 100% |
| Modal | 88.5 | 1.05s | 1.26s | 1.39s | 100% |
| Blaxel | 86.6 | 1.05s | 1.76s | 1.81s | 100% |
| Declaw | 88.3 | 1.10s | 1.28s | 1.30s | 100% |
| Mosaic | 87.8 | 1.17s | 1.27s | 1.35s | 100% |
| E2B | 85.3 | 1.33s | 1.60s | 1.77s | 100% |
| Buddy | 85.3 | 1.39s | 1.58s | 1.62s | 100% |
| Freestyle | 84.1 | 1.44s | 1.78s | 1.85s | 100% |
| Tensorlake | 75.3 | 2.40s | 2.58s | 2.58s | 100% |
| Sail | 67.1 | 2.83s | 3.94s | 4.03s | 100% |
| Sandbox0 | 48.0 | 4.81s | 5.77s | 5.84s | 100% |
| Upstash | 20.0 | 6.78s | 9.79s | 9.88s | 100% |
| CodeSandbox | 9.7 | 8.38s | 11.49s | 12.00s | 100% |
| Run Cloud | 0.0 | 35.45s | 61.68s | 61.70s | 42% |
| Hopx | 0.0 | 0.00s | 0.00s | 0.00s | 0% |
| Microsandbox | 0.0 | 0.00s | 0.00s | 0.00s | 0% |
| Northflank | 0.0 | 0.00s | 0.00s | 0.00s | 0% |
Want to see a provider added?
Methodology
What We Measure
Every benchmark measures Time to Interactive (TTI) — the elapsed time from calling compute.sandbox.create() to the first successful runCommand() inside the sandbox.
Each provider is tested with 100 iterations per run. Benchmarks run automatically via GitHub Actions on a recurring schedule. All results are committed to the public benchmarks repo.
Burst Test: All sandboxes are launched concurrently in a single burst.
How We Score
The Composite Score is a weighted blend of timing metrics multiplied by the success rate. Each metric is scored against a fixed 10-second ceiling: 100 × (1 − value / 10,000ms), so a 200ms median scores 98 and anything ≥10s scores 0.
The weighted timing score is then multiplied by the success rate (0–1), so providers that fail frequently are penalized proportionally.
- • Median: 60% — primary signal for typical experience
- • P95: 25% — tail latency / consistency
- • P99: 15% — extreme tail latency
Sandbox Benchmarks FAQs
Have another question? Email us.
A sandbox is anywhere you can run code in isolation. It could be a VM, bare metal, a container, anywhere with compute resources.