Geometric Reasoning on GeoQA mid-school level
79.58Accuracy (GeoQA mid-school)Gemini-2.5-pro
Evaluation Results
| Method | Links | |
|---|---|---|
| Gemini-2.5-proRelease Date=Jun, 20252025.04 | 79.58 | |
| OpenAI-o3Release Date=Apr, 20252025.04 | 79.31 | |
| Qwen2.5-VL-72BRelease Date=Jan, 20252025.04 | 67.9 | |
| Intern-S1Release Date=Jul, 20252025.04 | 62.47 | |
| Claude-3.7-sonnetRelease Date=Feb, 20252025.04 | 49.73 | |
| GPT-4oRelease Date=May, 20242025.04 | 42.31 |