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 | 0.3599 | $12.71 | 725.7M | 1.6M | 10h 38m | 1614 | 4 |
| 2 | DeepSeek V4 Flash Vision Exp | 0.3164 | $9.48 | 551.4M | 1.2M | 7h 58m | 1228 | 8 |
| 3 | DeepSeek V4 Flash Vision Exp | 0.2513 | $13.44 | 768.2M | 1.8M | 13h 43m | 1760 | 7 |
| 4 | DeepSeek V4 Flash Vision Exp | 0.0143 | $12.56 | 761.2M | 1.2M | 12h 33m | 1627 | 35 |
| 5 | DeepSeek V4 Flash Vision Exp | 0.0000 | $11.64 | 694.6M | 1.3M | 11h 22m | 1413 | 33 |