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 | 0.6319 | $2653.82 | 1245.9M | 2.6M | 19h 52m | 2769 | 39 |
| 2 | GPT-6 Astra | 0.6316 | $2364.83 | 1112.0M | 2.4M | 19h 31m | 2539 | 51 |
| 3 | GPT-6 Astra | 0.6308 | $1877.69 | 889.9M | 1.7M | 16h 9m | 1997 | 31 |
| 4 | GPT-6 Astra | 0.6304 | $348.12 | 162.7M | 418k | 3h 11m | 442 | 4 |
| 5 | GPT-6 Astra | 0.6223 | $321.27 | 149.5M | 395k | 3h 5m | 401 | 4 |