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 | 0.2915 | $236.87 | 396.3M | 962k | 16h 44m | 955 | 7 |
| 2 | Claude Opus 5 | 0.1550 | $162.82 | 258.1M | 858k | 16h 52m | 679 | 10 |
| 3 | Claude Opus 5 | 0.1213 | $221.80 | 357.8M | 1.1M | 16h 0m | 789 | 12 |
| 4 | Claude Opus 5 | 0.0902 | $217.40 | 355.4M | 1.0M | 14h 55m | 890 | 16 |
| 5 | Claude Opus 5 | 0.0000 | $223.36 | 369.6M | 980k | 18h 26m | 878 | 18 |