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 | 0.9586 | $14.22 | 21.7M | 140k | 16h 34m | 179 | 11 |
| 2 | Claude Fable 5.1 | 0.9369 | $12.11 | 18.4M | 119k | 17h 11m | 175 | 15 |
| 3 | Claude Fable 5.1 | 0.9113 | $11.26 | 16.5M | 124k | 17h 50m | 150 | 5 |
| 4 | Claude Fable 5.1 | 0.8820 | $17.05 | 26.5M | 157k | 16h 32m | 179 | 7 |
| 5 | Claude Fable 5.1 | 0.8229 | $11.99 | 18.2M | 119k | 18h 9m | 171 | 13 |