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 | 0.9740 | $14.41 | 21.9M | 144k | 17h 0m | 153 | 5 |
| 2 | Claude Fable 5.1 | 0.8920 | $12.71 | 19.5M | 122k | 16h 12m | 146 | 4 |
| 3 | Claude Fable 5.1 | 0.7800 | $16.40 | 25.5M | 151k | 15h 15m | 165 | 3 |
| 4 | Claude Fable 5.1 | 0.6180 | $15.24 | 23.4M | 145k | 17h 43m | 162 | 6 |
| 5 | Claude Fable 5.1 | 0.4080 | $13.04 | 19.3M | 141k | 15h 46m | 135 | 4 |