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 | 0.9298 | $976.76 | 847.7M | 1.7M | 19h 7m | 2003 | 24 |
| 2 | Claude Fable 5 | 0.9264 | $317.64 | 267.5M | 691k | 8h 11m | 499 | 15 |
| 3 | Claude Fable 5 | 0.9178 | $1055.99 | 927.6M | 1.7M | 18h 46m | 2232 | 22 |
| 4 | Claude Fable 5 | 0.9144 | $317.29 | 267.0M | 700k | 14h 26m | 483 | 17 |
| 5 | Claude Fable 5 | 0.8973 | $914.45 | 785.4M | 1.7M | 18h 23m | 1843 | 35 |