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 | 0.5001 | $197.63 | 316.4M | 1.0M | 15h 45m | 730 | 24 |
| 2 | Claude Opus 5 | 0.4607 | $210.30 | 340.0M | 1.1M | 18h 37m | 821 | 19 |
| 3 | Claude Opus 5 | 0.4307 | $172.30 | 276.6M | 886k | 18h 12m | 665 | 10 |
| 4 | Claude Opus 5 | 0.2573 | $170.72 | 274.5M | 864k | 18h 25m | 685 | 12 |
| 5 | Claude Opus 5 | 0.0762 | $115.37 | 188.1M | 559k | 19h 9m | 407 | 11 |