Geometric Problem Solving on GeoTrust Tier4 (test)
23CountIntern-S1
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Intern-S1#Params=235B+6B2025.04 | 23 | 38.33 | |
| OpenAI-o3#Params=-2025.04 | 23 | 38.33 | |
| Gemini-2.5-pro#Params=-2025.04 | 20 | 33.33 | |
| DeepSeek-R1#Params=671B, input_format=only natural language questions, no images2025.04 | 17 | 28.33 | |
| GPT-4o#Params=-2025.04 | 10 | 16.67 | |
| Claude-3.7-sonnet#Params=-2025.04 | 7 | 11.67 | |
| Qwen2.5-VL-72B#Params=72B2025.04 | 6 | 10 | |
| Qwen2-VL-7B#Params=7B2025.04 | 2 | 3.33 | |
| LLaVA-1.5-7B#Params=7B2025.04 | 0 | 0 |