AI Agent infrastructure, measured.
Independent benchmarks and teardowns of the infrastructure agents run on: sandboxes, agent VMs, computer-use and browser infra, and agent evals. Every number links to the raw run.
See the leaderboardWork with me
Time to first command
How long from asking for a fresh sandbox until a shell command actually runs in it.
Cold create to first command
ms · lower is better · p50 of 5 fresh sandboxes each
- 1Cloudflare Sandbox483 ms×1.00
- 2Fly.io Sprites623 ms×1.29
- 3Vercel Sandbox662 ms×1.37
- 4Freestyle976 ms×2.02
- 5Tensorlake1.92 s×3.97
- 6BoxLite Cloud4.07 s×8.42
- 7Upstash Box5.02 s×10.39
- 1Cloudflare Sandbox467 ms×1.00
- 2Fly.io Sprites505 ms×1.08
- 3Vercel Sandbox632 ms×1.35
- 4Freestyle897 ms×1.92
- 5Tensorlake1.90 s×4.06
- 6Upstash Box2.76 s×5.90
- 7BoxLite Cloud3.75 s×8.03
- 1Fly.io Sprites746 ms×1.00
- 2Cloudflare Sandbox888 ms×1.19
- 3Vercel Sandbox1.07 s×1.43
- 4Freestyle1.57 s×2.11
- 5Tensorlake2.49 s×3.34
- 6BoxLite Cloud4.43 s×5.94
- 7Upstash Box11.67 s×15.64
Method, raw results and the providers we could not measure →