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 | 0.9431 | $100.68 | 158.8M | 876k | 5h 35m | 408 | 11 |
| 2 | Claude Fable 5.1 | 0.9350 | $63.80 | 99.1M | 584k | 3h 4m | 218 | 8 |
| 3 | Claude Fable 5.1 | 0.9268 | $63.95 | 99.7M | 578k | 3h 23m | 222 | 5 |
| 4 | Claude Fable 5.1 | 0.9187 | $72.59 | 116.9M | 580k | 4h 7m | 245 | 5 |
| 5 | Claude Fable 5.1 | 0.9106 | $77.90 | 124.4M | 643k | 3h 55m | 256 | 4 |