List your company

Mathematical Problem Solving on FlowVerse Text Plus DI+RP+EI+OQ

41.7CoT ErrorQwen-VL-Plus
40.28849.81959.3568.881Mar 19, 2025

Evaluation results

MethodLinks
Qwen-VL-Plus2025.0341.735.6
InfiMM-Math-7B2025.0346.140.1
Gemini-Pro2025.0347.239.8
InternVL2.5-8B2025.0349.642.7
Qwen-VL-Max2025.0353.951
GPT-4o-mini2025.0359.655.2
Qwen2.5-VL-7B2025.036255
GPT-4o2025.0362.258.2
Qwen2-VL-72B2025.0363.750
Claude-3.5-Sonnet2025.036458.3
MathFlowBackbone=Qwen2.5-VL-7B2025.0364.256
VLM-R1-7B†2025.0365.558.9
InternVL2.5-78B2025.0367.864.7
Gemini-2.5-pro2025.0368.964.1
GPT-4V2025.037261.4
GPT-52025.037769.2
MathFlowBackbone=GPT-52025.037769.3