P95 stands for the 95th percentile. 95% of iterations completed at or below this time. Tells you the 'typical worst case' — what a user experiences on a bad-but-not-extreme run.

P99 stands for the 99th percentile. 99% of iterations completed at or below this time. It highlights the tail latency (the rare outlier spikes). With 100 iterations, this is the 99th value (second-worst)

Agents, sandboxes, & vibe-coded apps are powered by Cloud Run.

Sandbox Benchmarks

Q: What is Time to Interactive (TTI)?

TTI is the total elapsed time it takes for a sandbox provider to boot up a sandbox and run a terminal command inside the sandbox.

Q: Why use median instead of average?

The average can be easily skewed by a single extreme outlier. The median provides a more accurate representation of the typical performance most users will experience.

A leaderboard of common benchmarks for each of our sandbox providers.

Namespace· 4 vCPU / 16GB RAM· Northern Virginia, US

View on Github

Last run: September 4, 2026

Methodology

Performance Over Time

Composite Score

Detailed Metrics

Provider	Score	Median	P95	P99	Success
Isorun	99.4	0.06s	0.07s	0.07s	100%
CreateOS	98.2	0.17s	0.18s	0.18s	100%
OpenComputer	1.0	0.23s	0.23s	0.23s	1%
Declaw	96.4	0.31s	0.42s	0.46s	100%
Archil	95.2	0.45s	0.53s	0.54s	100%
Arker	94.1	0.56s	0.64s	0.65s	100%
Beam	93.2	0.60s	0.80s	0.80s	100%
Vercel	90.9	0.66s	1.05s	1.68s	100%
Miosa	92.6	0.70s	0.79s	0.82s	100%
Tenki	91.8	0.80s	0.86s	0.86s	100%
Modal	90.5	0.88s	0.98s	1.18s	100%
Superserve	88.1	1.00s	1.44s	1.54s	100%
Freestyle	87.7	1.11s	1.40s	1.43s	100%
Runloop	87.4	1.24s	1.29s	1.30s	100%
E2B	82.5	1.55s	1.98s	2.16s	100%
Tensorlake	79.7	1.97s	2.10s	2.12s	100%
Sail	73.7	2.49s	2.84s	2.88s	100%
Cloud Run	64.8	2.79s	4.49s	4.82s	100%
Mosaic	53.5	4.51s	4.74s	5.02s	100%
Upstash	40.4	4.82s	7.50s	7.95s	100%
Cloudflare	41.1	5.18s	6.31s	6.59s	95%
givemeanode	45.5	5.28s	5.71s	5.72s	100%
Blaxel	22.9	6.83s	8.32s	8.35s	89%
CodeSandbox	19.2	6.98s	9.57s	10.01s	100%
Run Cloud	12.6	7.91s	19.02s	23.25s	100%
Sandbox0	0.0	12.66s	12.75s	12.75s	5%
Daytona	0.0	44.45s	49.13s	49.13s	12%
Microsandbox	0.0	61.93s	110.65s	115.78s	94%

Want to see a provider added?

Let us know on X

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