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 | 0.6214 | $15.11 | 24.2M | 125k | 16h 58m | 204 | 11 |
| 2 | Claude Fable 5.1 | 0.6178 | $12.69 | 20.4M | 104k | 18h 1m | 175 | 12 |
| 3 | Claude Fable 5.1 | 0.6011 | $16.81 | 27.4M | 129k | 16h 5m | 208 | 16 |
| 4 | Claude Fable 5.1 | 0.5960 | $16.06 | 25.6M | 135k | 16h 36m | 193 | 12 |
| 5 | Claude Fable 5.1 | 0.0000 | $14.60 | 23.4M | 121k | 14h 28m | 186 | 8 |