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 | 0.7140 | $29.77 | 41.4M | 245k | 19h 28m | 207 | 2 |
| 2 | Claude Opus 5 | 0.6840 | $35.58 | 51.9M | 259k | 19h 2m | 225 | 2 |
| 3 | Claude Opus 5 | 0.6660 | $32.36 | 44.1M | 277k | 18h 32m | 192 | 4 |
| 4 | Claude Opus 5 | 0.5280 | $40.04 | 56.7M | 315k | 17h 56m | 222 | 4 |
| 5 | Claude Opus 5 | 0.2740 | $33.23 | 45.6M | 283k | 19h 1m | 200 | 2 |