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 | GLM-5.3 Flash | 0.4031 | $18.46 | 598.3M | 640k | 20h 1m | 1200 | 1 |
| 2 | GLM-5.3 Flash | 0.4009 | $22.72 | 722.1M | 1.2M | 19h 26m | 1667 | 27 |
| 3 | GLM-5.3 Flash | 0.3666 | $15.53 | 455.8M | 658k | 20h 1m | 864 | 1 |
| 4 | GLM-5.3 Flash | 0.2155 | $15.10 | 395.1M | 560k | 20h 1m | 851 | 1 |
| 5 | GLM-5.3 Flash | 0.1396 | $17.07 | 519.3M | 626k | 20h 0m | 1015 | 1 |