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 | 0.3166 | $19.44 | 30.2M | 179k | 16h 28m | 214 | 8 |
| 2 | Claude Fable 5.1 | 0.3123 | $26.35 | 42.8M | 204k | 16h 51m | 240 | 12 |
| 3 | Claude Fable 5.1 | 0.3116 | $13.99 | 21.5M | 133k | 17h 11m | 192 | 11 |
| 4 | Claude Fable 5.1 | 0.3057 | $16.57 | 26.2M | 144k | 13h 23m | 202 | 11 |
| 5 | Claude Fable 5.1 | 0.2155 | $25.98 | 41.7M | 212k | 14h 56m | 237 | 10 |