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 | 0.9536 | $56.70 | 98.3M | 311k | 9h 23m | 314 | 7 |
| 2 | Claude Fable 5.1 | 0.9536 | $46.08 | 78.9M | 273k | 10h 59m | 276 | 9 |
| 3 | Claude Fable 5.1 | 0.9429 | $31.38 | 51.1M | 239k | 11h 53m | 233 | 6 |
| 4 | Claude Fable 5.1 | 0.8357 | $41.84 | 71.9M | 245k | 9h 32m | 252 | 11 |
| 5 | Claude Fable 5.1 | 0.7929 | $38.22 | 64.0M | 257k | 7h 37m | 245 | 4 |