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 | 1.0000 | $34.17 | 49.9M | 378k | 2h 37m | 164 | 6 |
| 2 | Claude Fable 5.1 | 1.0000 | $45.30 | 70.2M | 418k | 4h 0m | 200 | 9 |
| 3 | Claude Fable 5.1 | 1.0000 | $34.22 | 50.9M | 360k | 1h 59m | 159 | 8 |
| 4 | Claude Fable 5.1 | 1.0000 | $43.08 | 67.7M | 379k | 4h 3m | 206 | 11 |
| 5 | Claude Fable 5.1 | 1.0000 | $50.91 | 83.1M | 385k | 3h 9m | 232 | 8 |