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 | DeepSeek V4 Flash Vision Exp | 0.9788 | $23.06 | 1383.4M | 2.4M | 13h 19m | 2911 | 21 |
| 2 | DeepSeek V4 Flash Vision Exp | 0.2005 | $21.27 | 1250.2M | 2.4M | 15h 16m | 2676 | 17 |
| 3 | DeepSeek V4 Flash Vision Exp | 0.0000 | $11.39 | 676.4M | 1.3M | 7h 45m | 1507 | 15 |
| 4 | DeepSeek V4 Flash Vision Exp | 0.0000 | $22.43 | 1385.4M | 2.0M | 11h 31m | 2860 | 2 |
| 5 | DeepSeek V4 Flash Vision Exp | 0.0000 | $18.34 | 1059.5M | 2.2M | 17h 1m | 2383 | 9 |