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 | 0.9986 | $19.92 | 56.8M | 329k | 17h 53m | 217 | 9 |
| 2 | Claude Opus 5.5 | 0.9657 | $21.32 | 60.2M | 359k | 15h 52m | 219 | 7 |
| 3 | Claude Opus 5.5 | 0.9563 | $17.15 | 48.3M | 286k | 18h 37m | 196 | 8 |
| 4 | Claude Opus 5.5 | 0.9209 | $19.00 | 56.2M | 297k | 18h 34m | 231 | 12 |
| 5 | Claude Opus 5.5 | 0.4697 | $23.43 | 66.1M | 395k | 18h 40m | 232 | 8 |