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 3.8 Flash | 0.9006 | $5.19 | 42.9M | 142k | 3h 20m | 258 | 6 |
| 2 | Gemini 3.8 Flash | 0.8493 | $4.81 | 37.8M | 168k | 3h 17m | 232 | 5 |
| 3 | Gemini 3.8 Flash | 0.0997 | $3.73 | 23.7M | 117k | 2h 23m | 194 | 6 |
| 4 | Gemini 3.8 Flash | 0.0439 | $5.15 | 42.3M | 159k | 2h 44m | 260 | 3 |
| 5 | Gemini 3.8 Flash | 0.0009 | $42.37 | 149.3M | 250k | 10h 22m | 415 | 10 |