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 | Gemini 4 Carbon | 0.9268 | $112.64 | 321.3M | 1.4M | 19h 55m | 763 | 11 |
| 2 | Gemini 4 Carbon | 0.8907 | $78.20 | 248.2M | 1.1M | 13h 8m | 515 | 11 |
| 3 | Gemini 4 Carbon | 0.8842 | $58.51 | 166.7M | 956k | 13h 14m | 408 | 8 |
| 4 | Gemini 4 Carbon | 0.8825 | $61.99 | 196.8M | 1.1M | 12h 30m | 439 | 7 |
| 5 | Gemini 4 Carbon | 0.8458 | $58.27 | 224.6M | 1.0M | 11h 25m | 507 | 8 |