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 | 0.2139 | $1.97 | 101.8M | 387k | 13h 26m | 420 | 1 |
| 2 | DeepSeek V4 Flash Vision Exp | 0.2028 | $2.58 | 144.1M | 374k | 19h 46m | 508 | 2 |
| 3 | DeepSeek V4 Flash Vision Exp | 0.1948 | $3.01 | 169.4M | 449k | 11h 0m | 551 | 4 |
| 4 | DeepSeek V4 Flash Vision Exp | 0.1939 | $2.93 | 164.0M | 435k | 14h 31m | 526 | 4 |
| 5 | DeepSeek V4 Flash Vision Exp | 0.1818 | $2.40 | 126.7M | 415k | 10h 56m | 421 | 6 |