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 | 0.8458 | $138.31 | 228.0M | 1000k | 9h 1m | 580 | 15 |
| 2 | Claude Fable 5.1 | 0.8393 | $189.96 | 320.0M | 1.2M | 8h 30m | 718 | 28 |
| 3 | Claude Fable 5.1 | 0.8171 | $168.52 | 281.0M | 1.2M | 8h 8m | 667 | 16 |
| 4 | Claude Fable 5.1 | 0.8047 | $204.26 | 349.5M | 1.2M | 8h 27m | 784 | 31 |
| 5 | Claude Fable 5.1 | 0.7867 | $157.78 | 258.9M | 1.2M | 7h 40m | 582 | 11 |