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 | 0.9354 | $108.05 | 175.6M | 512k | 12h 46m | 386 | 16 |
| 2 | Claude Opus 5 | 0.9312 | $80.95 | 128.6M | 410k | 11h 23m | 326 | 7 |
| 3 | Claude Opus 5 | 0.9268 | $66.75 | 102.7M | 380k | 7h 37m | 277 | 3 |
| 4 | Claude Opus 5 | 0.9178 | $120.16 | 201.7M | 487k | 16h 10m | 440 | 12 |
| 5 | Claude Opus 5 | 0.9082 | $82.70 | 130.5M | 438k | 8h 38m | 327 | 9 |