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 | 0.2272 | $15.45 | 22.4M | 175k | 11h 13m | 164 | 5 |
| 2 | Claude Fable 5.1 | 0.1750 | $17.15 | 26.3M | 165k | 12h 46m | 182 | 6 |
| 3 | Claude Fable 5.1 | 0.1459 | $14.90 | 22.4M | 152k | 15h 9m | 171 | 4 |
| 4 | Claude Fable 5.1 | 0.1343 | $11.08 | 15.3M | 141k | 13h 28m | 134 | 3 |
| 5 | Claude Fable 5.1 | 0.0226 | $15.74 | 21.7M | 202k | 14h 5m | 141 | 4 |