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 | 0.3591 | $41.25 | 57.4M | 336k | 16h 38m | 244 | 11 |
| 2 | Claude Opus 5 | 0.3182 | $39.66 | 57.3M | 295k | 17h 37m | 258 | 3 |
| 3 | Claude Opus 5 | 0.3001 | $47.82 | 68.3M | 371k | 18h 4m | 251 | 14 |
| 4 | Claude Opus 5 | 0.1306 | $32.65 | 45.8M | 261k | 17h 4m | 232 | 9 |
| 5 | Claude Opus 5 | 0.0000 | $41.18 | 59.8M | 304k | 17h 59m | 267 | 9 |