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 | $3.36 | 53.3M | 330k | 10h 44m | 230 | 6 |
| 2 | GLM-5.3 Flash | 0.0000 | $3.57 | 84.9M | 393k | 10h 21m | 284 | 2 |
| 3 | GLM-5.3 Flash | 0.0000 | $2.05 | 35.6M | 271k | 7h 4m | 202 | 3 |
| 4 | GLM-5.3 Flash | 0.0000 | $1.70 | 38.2M | 299k | 3h 32m | 183 | 4 |
| 5 | GLM-5.3 Flash | 0.0000 | $4.90 | 95.4M | 391k | 15h 5m | 312 | 7 |