List your company

Multimodal Mathematical Reasoning on FlowVerse Vision Dense

28.8CoT ErrorInfiMM-Math-7B
27.62435.56243.551.438Mar 19, 2025

Evaluation results

MethodLinks
InfiMM-Math-7BModel Type=Open-source MLLMs2025.0328.815.4
VLM-R1-7B†Model Type=Open-source MLLMs2025.0336.224.5
Math-LLaVA-13BModel Type=Math-specialized MLLMs2025.0336.218.6
InternVL2.5-8BModel Type=Open-source MLLMs2025.0338.420.2
Qwen-VL-MaxModel Type=Closed-source MLLMs2025.0338.615.2
Qwen2-VL-72BModel Type=Open-source MLLMs2025.0340.825.3
GPT-4o-miniModel Type=Closed-source MLLMs2025.0341.126
MultiMath-7BModel Type=Math-specialized MLLMs2025.0341.722.1
SVE-Math-Qwen2.5-7BModel Type=Math-specialized MLLMs2025.0344.228.6
Qwen2.5-VL-7BModel Type=Open-source MLLMs2025.034531
Claude-3.5-SonnetModel Type=Closed-source MLLMs2025.034525.4
GPT-4oModel Type=Closed-source MLLMs2025.0345.230
GPT-4VModel Type=Closed-source MLLMs2025.0348.130.3
InternVL2.5-78BModel Type=Open-source MLLMs2025.0348.734.3
MathFlowModel Type=Open-source MLLMs, Backbone=Qwen2.5-VL-7B2025.034939.1
Gemini-2.5-proModel Type=Closed-source MLLMs2025.0352.137.1
GPT-5Model Type=Closed-source MLLMs2025.0353.844.7
MathFlowModel Type=Closed-source MLLMs, Backbone=GPT-52025.0358.254.1