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 | 0.3856 | $635.16 | 1128.1M | 1.8M | 19h 53m | 2411 | 22 |
| 2 | Claude Opus 5 | 0.3683 | $533.23 | 935.7M | 1.7M | 19h 23m | 1826 | 15 |
| 3 | Claude Opus 5 | 0.3411 | $614.97 | 1098.7M | 1.7M | 19h 37m | 2206 | 22 |
| 4 | Claude Opus 5 | 0.3150 | $429.99 | 753.2M | 1.4M | 19h 44m | 1660 | 18 |
| 5 | Claude Opus 5 | 0.3034 | $490.76 | 870.1M | 1.4M | 19h 44m | 1974 | 13 |