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 | GPT-6 Astra | 0.9965 | $2541.33 | 1210.9M | 2.5M | 19h 50m | 2830 | 27 |
| 2 | GPT-6 Astra | 0.9845 | $2685.88 | 1278.3M | 2.6M | 20h 0m | 2961 | 38 |
| 3 | GPT-6 Astra | 0.9844 | $2735.71 | 1297.4M | 2.6M | 19h 42m | 2884 | 29 |
| 4 | GPT-6 Astra | 0.9841 | $2589.72 | 1228.2M | 2.6M | 19h 59m | 2801 | 34 |
| 5 | GPT-6 Astra | 0.9841 | $2728.19 | 1288.8M | 2.7M | 19h 57m | 2941 | 47 |