Geometric Reasoning on OlympiadBench-Geo olympiad level
77.68AccuracyOpenAI-o3
Evaluation Results
| Method | Links | |
|---|---|---|
| OpenAI-o3Release Date=Apr, 20252025.04 | 77.68 | |
| Gemini-2.5-proRelease Date=Jun, 20252025.04 | 75 | |
| Intern-S1Release Date=Jul, 20252025.04 | 49.11 | |
| Qwen2.5-VL-72BRelease Date=Jan, 20252025.04 | 29.46 | |
| Claude-3.7-sonnetRelease Date=Feb, 20252025.04 | 17.86 | |
| GPT-4oRelease Date=May, 20242025.04 | 13.39 |