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 | 0.5368 | $28.63 | 38.3M | 258k | 18h 34m | 202 | 8 |
| 2 | Claude Opus 5 | 0.5328 | $33.95 | 47.5M | 267k | 18h 38m | 217 | 7 |
| 3 | Claude Opus 5 | 0.5241 | $22.96 | 29.3M | 223k | 18h 54m | 175 | 5 |
| 4 | Claude Opus 5 | 0.4833 | $26.82 | 34.5M | 255k | 17h 35m | 181 | 3 |
| 5 | Claude Opus 5 | 0.2240 | $27.38 | 35.8M | 258k | 19h 9m | 186 | 9 |