OpenTable
Rebuilt by Claude Opus 5.5: 150 of 512 hidden tests passed (29.3%)
reference.local/
Referencerebuild.local/
Claude Opus 5.5| Model | Passed | % | Screenshot |
|---|---|---|---|
| 150 | 29.3 | ||
| 94 | 18.4 | ||
| 93 | 18.2 | ||
| 82 | 16.0 | ||
| 81 | 15.8 | ||
| 79 | 15.4 | ||
| 35 | 6.8 | ||
| 0 | 0.0 | — |