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 | $174.84 | 280.2M | 919k | 15h 16m | 714 | 20 |
| 2 | Claude Opus 5 | 1.0000 | $204.51 | 327.4M | 1.1M | 13h 23m | 746 | 19 |
| 3 | Claude Opus 5 | 0.9984 | $103.75 | 161.1M | 640k | 5h 43m | 315 | 4 |
| 4 | Claude Opus 5 | 0.9841 | $206.83 | 330.5M | 1.1M | 10h 19m | 761 | 12 |
| 5 | Claude Opus 5 | 0.9635 | $169.99 | 269.9M | 940k | 17h 12m | 655 | 17 |