Share your thoughts, 1 month free Claude Pro on usSee more

Mathematical Problem Solving on MathVerse (testmini)

64.9Accuracy

Human

Updated 1mo ago

Evaluation Results

Method	Links
Human 2025.05		64.9
Gemini-1.5-Pro 2025.05		51.1
GPT-4o 2025.05		50.8
Claude3.5-Sonnet 2025.05		49
MathCoder-VL-8B 2025.05		46.5
MathGLM-Vision-9B 2025.05		44.2
GPT-4-turbo 2025.05		43.5
InternVL2-76B 2025.05		42.8
InternVL2-26B 2025.05		40
GPT-4V 2025.05		39.4
Qwen-VL-Max 2025.05		35.9
InternVL2-8B 2025.05		35.9
MathCoder-VL-2B 2025.05		35.4
MAVIS-7B 2025.05		35.2
Qwen2-VL 2025.05		33.6
Math-PUMA-Qwen2 2025.05		33.6
Claude3-Opus 2025.05		31.8
Math-PUMA-DS 2025.05		31.8
Multimath-7B 2025.05		27.7
IXC-2-VL 2025.05		25.9
InternVL-Chat-2B-V1-5 2025.05		23.1
Math-LLaVA-13B 2025.05		22.9
Qwen-VL-Plus 2025.05		21.3
Deepseek-VL 2025.05		19.3
IXC-2.5-Reward 2025.05		18.8
SPHINX-V2-13B 2025.05		16.1
LLaVA-1.5-13B 2025.05		12.7
Random Chance 2025.05		12.4