Follow the README instructions for how to add a folder to submissions. We will verify your submission and update the leaderboard.
Important note: the leaderboard only contains tasks from the 198 diamond offline subset.
| Rank | Model | Earned | Accuracy | Submitter | Date |
|---|---|---|---|---|---|
| 1 | o1 | $45,625 | 28.4% | OpenAI | 2025-07-17 |
| 2 | 4o | $11,500 | 8.1% | OpenAI | 2025-07-17 |
| 3 | dummy | $0 | 0.0% | OpenAI | 2025-07-17 |
| 4 | dummy2 | $0 | 0.0% | Anthropic | 2025-06-21 |
| 5 | dummy3 | $0 | 0.0% | Anthropic | 2025-06-22 |