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 | 0.0316 | $3.90 | 73.7M | 369k | 19h 44m | 299 | 7 |
| 2 | GLM-5.3 Flash | 0.0303 | $10.53 | 223.0M | 596k | 19h 20m | 502 | 15 |
| 3 | GLM-5.3 Flash | 0.0288 | $2.96 | 83.8M | 329k | 18h 56m | 365 | 2 |
| 4 | GLM-5.3 Flash | 0.0270 | $4.30 | 96.9M | 346k | 19h 28m | 348 | 3 |
| 5 | GLM-5.3 Flash | 0.0056 | $7.03 | 151.5M | 495k | 17h 26m | 374 | 10 |