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 | 0.0569 | $14.19 | 350.7M | 1.3M | 17h 49m | 793 | 3 |
| 2 | GLM-5.3 Flash | 0.0081 | $17.03 | 435.1M | 1.2M | 17h 59m | 1065 | 16 |
| 3 | GLM-5.3 Flash | 0.0000 | $10.99 | 299.0M | 1.2M | 12h 18m | 713 | 5 |
| 4 | GLM-5.3 Flash | 0.0000 | $9.30 | 218.2M | 1.0M | 15h 25m | 600 | 4 |
| 5 | GLM-5.3 Flash | 0.0000 | $7.12 | 187.9M | 1.1M | 14h 19m | 562 | 4 |