External Tester Review, November 2, 2026
November 2, 2026
An outside tester paid through x402 and ran a live H100 job. We checked every step: payment, machine order, boot, health checks and shutdown. Read the results.
Reports
Published weekly. Newest first. Each report includes a summary and the full PDF.
Live launch and ready rate counts: Capacity IndexHow we test
Method
Controlled host failure drills, with the timings published.
On Oct. 14, 2026 we conducted a controlled failover drill on an internal Kilawatt test account. A GPU workload wrote a checkpoint roughly once per minute. We then terminated its host machine at the supplier mid run and measured the time to detect the failure, the time to provision a replacement on a different supplier, and the time to resume from the last checkpoint.
| Run | Path | Time to detect | Time to provision | Time to resume | Total recovery | Result |
|---|---|---|---|---|---|---|
| Run 1 | RunPod to Hyperstack | 238 s | 239 s | 175 s (manual) | 10.9 min | Failure detected, replacement provisioned, checkpoint intact. Resume was performed manually because of a defect in our test script, since corrected. |
| Run 2 | RunPod to Vast.ai | 121 s | 119 s | 12 s (automatic) | 4.2 min | Failure detected, replacement provisioned, resumed from checkpoint 4 automatically. |
| Excluded run | Vast.ai, no fault injected | n/a | n/a | n/a | n/a | Stopped before the drill. Machine boot took 14.6 minutes, consuming most of the 20-minute job window. Not a test result. |
Kilawatt detected both host failures without intervention and provisioned the workload again on a different supplier. The latest checkpoint was intact on the replacement in both runs. In run 2 the workload resumed from checkpoint 4 automatically and went on to write checkpoints 5, 6 and 7.