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 | 0.0000 | $219.62 | 392.8M | 963k | 7h 58m | 844 | 9 |
| 2 | Claude Fable 5.1 | 0.0000 | $193.70 | 342.2M | 936k | 9h 59m | 820 | 11 |
| 3 | Claude Fable 5.1 | 0.0000 | $177.26 | 316.2M | 794k | 11h 11m | 833 | 9 |
| 4 | Claude Fable 5.1 | 0.0000 | $182.53 | 329.2M | 746k | 9h 31m | 841 | 21 |
| 5 | Claude Fable 5.1 | 0.0000 | $226.14 | 406.3M | 953k | 11h 14m | 902 | 15 |