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 Fable 5.1 | 0.8857 | $37.59 | 60.9M | 295k | 9h 18m | 239 | 6 |
| 2 | Claude Fable 5.1 | 0.8752 | $40.65 | 67.2M | 290k | 7h 39m | 274 | 9 |
| 3 | Claude Fable 5.1 | 0.8680 | $40.94 | 67.5M | 295k | 10h 23m | 279 | 9 |
| 4 | Claude Fable 5.1 | 0.8599 | $41.23 | 67.7M | 303k | 9h 29m | 256 | 9 |
| 5 | Claude Fable 5.1 | 0.8541 | $59.53 | 101.1M | 368k | 11h 5m | 353 | 7 |