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 | 0.5674 | $350.25 | 628.4M | 1.5M | 16h 3m | 1456 | 82 |
| 2 | Claude Fable 5.1 | 0.5395 | $313.80 | 554.2M | 1.5M | 11h 26m | 1289 | 34 |
| 3 | Claude Fable 5.1 | 0.4885 | $375.68 | 676.2M | 1.5M | 17h 3m | 1505 | 50 |
| 4 | Claude Fable 5.1 | 0.4469 | $289.07 | 503.7M | 1.5M | 13h 41m | 1156 | 50 |
| 5 | Claude Fable 5.1 | 0.2880 | $179.04 | 308.9M | 1.0M | 10h 29m | 755 | 12 |