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 Opus 5 | 0.9977 | $235.26 | 378.0M | 1.2M | 15h 3m | 814 | 17 |
| 2 | Claude Opus 5 | 0.8852 | $226.31 | 360.9M | 1.2M | 15h 56m | 756 | 18 |
| 3 | Claude Opus 5 | 0.8705 | $196.81 | 310.4M | 1.1M | 15h 31m | 680 | 10 |
| 4 | Claude Opus 5 | 0.8628 | $231.46 | 372.7M | 1.2M | 17h 30m | 794 | 16 |
| 5 | Claude Opus 5 | 0.6631 | $126.31 | 185.3M | 890k | 10h 20m | 498 | 10 |