Every run: the agent's full trajectory, its submissions over the 20-hour budget, and the final score.
| # | Model | Reward | Cost | In tok | Out tok | Time | Steps | Submits |
|---|---|---|---|---|---|---|---|---|
| 1 | Claude Opus 5.5 | 0.9929 | $30.52 | 100.2M | 393k | 11h 17m | 272 | 6 |
| 2 | Claude Opus 5.5 | 0.9786 | $45.44 | 160.9M | 500k | 13h 11m | 377 | 11 |
| 3 | Claude Opus 5.5 | 0.9786 | $34.33 | 116.2M | 417k | 11h 0m | 318 | 8 |
| 4 | Claude Opus 5.5 | 0.9750 | $38.61 | 139.5M | 396k | 12h 25m | 373 | 6 |
| 5 | Claude Opus 5.5 | 0.9571 | $39.47 | 142.7M | 408k | 13h 40m | 381 | 7 |