List your company

Geometric Problem Solving on GeoTrust Tier1 (test)

37CountOpenAI-o3
1.6410.822029.18Apr 22, 2025

Evaluation results

MethodLinks
OpenAI-o3#Params=-2025.043761.67
Intern-S1#Params=235B+6B2025.043558.33
Gemini-2.5-pro#Params=-2025.043456.67
Claude-3.7-sonnet#Params=-2025.043355
Qwen2.5-VL-72B#Params=72B2025.043253.33
GPT-4o#Params=-2025.043151.67
DeepSeek-R1#Params=671B, input_format=only natural language questions, no images2025.043050
Qwen2-VL-7B#Params=7B2025.0458.33
LLaVA-1.5-7B#Params=7B2025.0435