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 | 1.0000 | $209.27 | 343.2M | 992k | 13h 38m | 826 | 10 |
| 2 | Claude Opus 5 | 1.0000 | $125.36 | 207.5M | 580k | 6h 45m | 426 | 9 |
| 3 | Claude Opus 5 | 1.0000 | $155.19 | 264.1M | 628k | 9h 49m | 494 | 11 |
| 4 | Claude Opus 5 | 0.9992 | $107.50 | 172.4M | 579k | 5h 52m | 375 | 4 |
| 5 | Claude Opus 5 | 0.9992 | $128.06 | 213.3M | 585k | 7h 58m | 454 | 2 |