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 | 0.5457 | $12.11 | 18.5M | 118k | 17h 49m | 156 | 8 |
| 2 | Claude Fable 5.1 | 0.5346 | $11.41 | 17.0M | 120k | 17h 44m | 158 | 7 |
| 3 | Claude Fable 5.1 | 0.5104 | $17.60 | 28.7M | 134k | 17h 45m | 230 | 11 |
| 4 | Claude Fable 5.1 | 0.4905 | $10.29 | 15.1M | 114k | 17h 40m | 140 | 7 |
| 5 | Claude Fable 5.1 | 0.4889 | $10.86 | 15.6M | 125k | 15h 34m | 138 | 6 |