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 | 0.0000 | $1.89 | 103.3M | 298k | 19h 46m | 435 | 3 |
| 2 | DeepSeek V4 Flash Vision Exp | 0.0000 | $2.64 | 142.6M | 438k | 12h 9m | 474 | 5 |
| 3 | DeepSeek V4 Flash Vision Exp | 0.0000 | $2.54 | 140.9M | 373k | 17h 33m | 473 | 5 |
| 4 | DeepSeek V4 Flash Vision Exp | 0.0000 | $2.76 | 151.1M | 447k | 15h 16m | 503 | 6 |
| 5 | DeepSeek V4 Flash Vision Exp | 0.0000 | $2.75 | 153.9M | 409k | 18h 4m | 510 | 2 |