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 | 0.8917 | $57.70 | 85.6M | 405k | 15h 40m | 277 | 6 |
| 2 | Claude Opus 5 | 0.8845 | $64.52 | 99.6M | 401k | 17h 6m | 313 | 3 |
| 3 | Claude Opus 5 | 0.8644 | $47.78 | 71.9M | 322k | 17h 28m | 282 | 6 |
| 4 | Claude Opus 5 | 0.8538 | $47.81 | 68.4M | 372k | 17h 11m | 247 | 7 |
| 5 | Claude Opus 5 | 0.8444 | $51.06 | 76.5M | 351k | 16h 37m | 268 | 3 |