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 | $8.45 | 217.9M | 542k | 13h 37m | 520 | 3 |
| 2 | GLM-5.3 Flash | 0.0000 | $10.25 | 278.8M | 597k | 10h 31m | 596 | 8 |
| 3 | GLM-5.3 Flash | 0.0000 | $13.39 | 288.1M | 552k | 17h 55m | 640 | 19 |
| 4 | GLM-5.3 Flash | 0.0000 | $7.94 | 248.0M | 455k | 19h 56m | 595 | 7 |
| 5 | GLM-5.3 Flash | 0.0000 | $12.92 | 278.9M | 847k | 19h 47m | 687 | 1 |