List your company

Mathematical Reasoning on FlowVerse Text Plus

46.1CoT ErrorInfiMM-Math-7B
44.86453.20761.5569.893Mar 19, 2025

Evaluation results

MethodLinks
InfiMM-Math-7BModel Type=Open-source MLLMs2025.0346.140.1
InternVL2.5-8BModel Type=Open-source MLLMs2025.0349.642.7
Qwen-VL-MaxModel Type=Closed-source MLLMs2025.0353.951
GPT-4o-miniModel Type=Closed-source MLLMs2025.0359.655.2
Qwen2.5-VL-7BModel Type=Open-source MLLMs2025.036255
GPT-4oModel Type=Closed-source MLLMs2025.0362.258.2
Qwen2-VL-72BModel Type=Open-source MLLMs2025.0363.750
Claude-3.5-SonnetModel Type=Closed-source MLLMs2025.036458.3
MathFlowModel Type=Open-source MLLMs, Backbone=Qwen2.5-VL-7B2025.0364.256
VLM-R1-7B†Model Type=Open-source MLLMs2025.0365.558.9
InternVL2.5-78BModel Type=Open-source MLLMs2025.0367.864.7
Gemini-2.5-proModel Type=Closed-source MLLMs2025.0368.964.1
GPT-4VModel Type=Closed-source MLLMs2025.037261.4
GPT-5Model Type=Closed-source MLLMs2025.037769.2
MathFlowModel Type=Closed-source MLLMs, Backbone=GPT-52025.037769.3