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 | 0.0779 | $45.19 | 77.6M | 262k | 17h 49m | 334 | 16 |
| 2 | Claude Fable 5.1 | 0.0754 | $52.35 | 88.6M | 331k | 12h 35m | 339 | 10 |
| 3 | Claude Fable 5.1 | 0.0711 | $71.87 | 125.0M | 385k | 17h 43m | 414 | 18 |
| 4 | Claude Fable 5.1 | 0.0695 | $54.46 | 93.0M | 327k | 16h 46m | 347 | 11 |
| 5 | Claude Fable 5.1 | 0.0678 | $47.98 | 81.1M | 304k | 16h 17m | 333 | 10 |