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 | 0.9772 | $29.82 | 38.2M | 293k | 18h 29m | 172 | 2 |
| 2 | Claude Opus 5 | 0.9539 | $25.73 | 31.4M | 274k | 19h 1m | 160 | 10 |
| 3 | Claude Opus 5 | 0.9168 | $37.68 | 49.7M | 354k | 18h 11m | 187 | 8 |
| 4 | Claude Opus 5 | 0.5927 | $30.96 | 39.3M | 312k | 17h 25m | 184 | 3 |
| 5 | Claude Opus 5 | 0.0000 | $33.24 | 46.7M | 272k | 18h 15m | 233 | 6 |