Runs
Every run of the benchmark, newest first. A full run covers every cloud, app and model; the others test a subset.
| Run | Scope | Models | Cells | Succeeded | Checks passed | Agent cost |
|---|---|---|---|---|---|---|
| Oct 14, 2026 · #9 | Full run | Opus 5.5 | 20 | 20 | 100% | $20.24 |
| Oct 13, 2026 · #8 | Full run | Opus 5.5 | 20 | 18 | 98% | $16.87 |
| Oct 13, 2026 · #5 | Apps: static, clouds: all, models: all | Opus 5.5 | 4 | 4 | 100% | $2.38 |
| Oct 13, 2026 · #4 | Apps: static, clouds: all, models: all | Opus 5.5 | 4 | 4 | 100% | $2.10 |