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 | 0.7029 | $6.19 | 134.2M | 441k | 12h 1m | 387 | 3 |
| 2 | GLM-5.3 Flash | 0.6977 | $3.78 | 63.8M | 292k | 10h 53m | 273 | 5 |
| 3 | GLM-5.3 Flash | 0.6919 | $5.75 | 129.8M | 427k | 15h 0m | 361 | 4 |
| 4 | GLM-5.3 Flash | 0.5673 | $4.20 | 81.9M | 369k | 16h 1m | 327 | 2 |
| 5 | GLM-5.3 Flash | 0.4693 | $5.24 | 113.1M | 401k | 10h 10m | 361 | 9 |