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 3.7 Flash | 0.1545 | $46.71 | 314.3M | 747k | 10h 28m | 740 | 2 |
| 2 | Gemini 3.7 Flash | 0.0976 | $71.79 | 423.8M | 777k | 17h 50m | 948 | 3 |
| 3 | Gemini 3.7 Flash | 0.0650 | $36.49 | 288.8M | 712k | 6h 10m | 725 | 2 |
| 4 | Gemini 3.7 Flash | 0.0000 | $51.00 | 393.0M | 804k | 11h 3m | 1002 | 4 |
| 5 | Gemini 3.7 Flash | 0.0000 | $24.96 | 238.1M | 690k | 3h 15m | 539 | 2 |