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 | 0.0237 | $8.27 | 504.6M | 723k | 9h 18m | 1260 | 11 |
| 2 | DeepSeek V4 Flash Vision Exp | 0.0107 | $5.01 | 303.8M | 439k | 7h 50m | 723 | 8 |
| 3 | DeepSeek V4 Flash Vision Exp | 0.0030 | $5.08 | 311.7M | 419k | 8h 3m | 781 | 6 |
| 4 | DeepSeek V4 Flash Vision Exp | 0.0000 | $3.68 | 220.8M | 344k | 8h 44m | 670 | 8 |
| 5 | DeepSeek V4 Flash Vision Exp | 0.0000 | $4.87 | 296.6M | 422k | 6h 43m | 730 | 3 |