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 | 0.9397 | $59.14 | 36.3M | 315k | 5h 54m | 162 | 8 |
| 2 | Claude Fable 5 | 0.9375 | $84.70 | 58.1M | 370k | 13h 38m | 228 | 14 |
| 3 | Claude Fable 5 | 0.2381 | $63.35 | 41.3M | 309k | 8h 51m | 197 | 11 |
| 4 | Claude Fable 5 | 0.2345 | $86.58 | 63.1M | 319k | 16h 11m | 264 | 7 |
| 5 | Claude Fable 5 | 0.1639 | $81.13 | 55.6M | 354k | 15h 53m | 224 | 11 |