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 | 0.4011 | $588.09 | 1056.3M | 2.5M | 17h 46m | 2515 | 13 |
| 2 | Claude Fable 5.1 | 0.3901 | $721.01 | 1319.3M | 2.5M | 17h 44m | 2872 | 14 |
| 3 | Claude Fable 5.1 | 0.3516 | $384.99 | 680.6M | 1.8M | 18h 59m | 1688 | 9 |
| 4 | Claude Fable 5.1 | 0.3352 | $475.86 | 850.2M | 2.1M | 18h 4m | 1894 | 7 |
| 5 | Claude Fable 5.1 | 0.2308 | $392.90 | 697.7M | 1.8M | 19h 16m | 1679 | 4 |