List your company

Plane Geometry Reasoning on PlaneFGeo3k

87.9Accuracy (%)human
14.68433.69252.771.708May 11, 2026

Evaluation results

MethodLinks
human2026.0587.9
Ground Truth2026.0584.1
Hilbert-Geoevaluation_samples=452026.0580.2
Gemini 2.5 Pro2026.0572.3
GPT-52026.0568.1
Claude 3.7 Sonnet2026.0560
Gemini 2.5 Flash2026.0555.6
GPT-4o2026.0547.2
Llama 3.2 70B2026.0533.3
Deepseek-V3 67B2026.0530.4
Qwen2.5-VL-Instruct-32B2026.0524.8
Qwen2.5-VL-Instruct-7B2026.0517.5