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 | 0.0000 | $22.01 | 646.4M | 1.1M | 19h 18m | 1514 | 1 |
| 2 | GLM-5.3 Flash | 0.0000 | $23.47 | 725.1M | 1.2M | 15h 11m | 1632 | 14 |
| 3 | GLM-5.3 Flash | 0.0000 | $28.19 | 886.8M | 1.2M | 17h 12m | 2083 | 25 |
| 4 | GLM-5.3 Flash | 0.0000 | $25.76 | 811.2M | 1.4M | 20h 0m | 2421 | 7 |
| 5 | GLM-5.3 Flash | 0.0000 | $21.51 | 630.1M | 971k | 19h 59m | 1458 | 2 |