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 | 0.8080 | $283.04 | 487.0M | 1.6M | 13h 9m | 1070 | 46 |
| 2 | Claude Fable 5.1 | 0.5509 | $202.06 | 343.4M | 1.3M | 9h 13m | 819 | 28 |
| 3 | Claude Fable 5.1 | 0.5106 | $148.93 | 254.0M | 906k | 6h 11m | 641 | 18 |
| 4 | Claude Fable 5.1 | 0.1368 | $219.11 | 385.2M | 1.1M | 10h 27m | 835 | 22 |
| 5 | Claude Fable 5.1 | 0.0000 | $217.24 | 363.8M | 1.5M | 10h 22m | 893 | 27 |