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 | $4.02 | 226.3M | 587k | 18h 16m | 503 | 5 |
| 2 | DeepSeek V4 Flash Vision Exp | 0.0000 | $4.14 | 229.6M | 637k | 20h 0m | 483 | 1 |
| 3 | DeepSeek V4 Flash Vision Exp | 0.0000 | $3.11 | 165.3M | 552k | 9h 15m | 409 | 6 |
| 4 | DeepSeek V4 Flash Vision Exp | 0.0000 | $3.44 | 189.2M | 548k | 10h 29m | 453 | 4 |
| 5 | DeepSeek V4 Flash Vision Exp | 0.0000 | $3.67 | 205.7M | 547k | 18h 16m | 510 | 3 |