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 | $3.26 | 101.7M | 249k | 9h 20m | 478 | 7 |
| 2 | GLM-5.3 Flash | 0.0000 | $5.76 | 124.8M | 287k | 14h 48m | 473 | 3 |
| 3 | GLM-5.3 Flash | 0.0000 | $3.52 | 102.4M | 346k | 8h 20m | 418 | 2 |
| 4 | GLM-5.3 Flash | 0.0000 | $5.38 | 133.7M | 325k | 9h 34m | 541 | 2 |
| 5 | GLM-5.3 Flash | 0.0000 | $5.24 | 138.3M | 282k | 10h 25m | 542 | 3 |