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 | 0.0249 | $124.87 | 207.2M | 534k | 12h 23m | 460 | 7 |
| 2 | Claude Opus 5 | 0.0223 | $101.43 | 161.0M | 523k | 12h 35m | 390 | 8 |
| 3 | Claude Opus 5 | 0.0048 | $110.39 | 179.3M | 515k | 12h 40m | 431 | 7 |
| 4 | Claude Opus 5 | 0.0000 | $89.33 | 141.7M | 467k | 15h 55m | 376 | 11 |
| 5 | Claude Opus 5 | 0.0000 | $158.96 | 248.3M | 890k | 16h 5m | 693 | 15 |