List your company
ResearchFormerly WizWand

Multimodal Evaluation on MME Chinese

60.37Overall ScoreFull Prec
1.329216.657131.98547.3129Aug 5, 2025
Updated 7mo ago

Evaluation results

MethodLinks
Full PrecModel=Qwen2.5-VL-32B-Instruct, Bit-width=Full Precision2025.0860.37
Full PrecModel=Qwen2-VL-7B-Instruct, Bit-width=Full Precision2025.0856.09
Full PrecModel=LLaVA-OneVision-7B, Bit-width=Full Precision2025.0853.93
Full PrecModel=Qwen2.5-VL-7B-Instruct, Bit-width=Full Precision2025.0852.98
GPTQModel=Qwen2.5-VL-32B-Instruct, Bit-width=3-bit2025.0847.76
VLMQModel=Qwen2-VL-7B-Instruct, Quantization Bit=INT42025.0847.73
GPTAQModel=Qwen2-VL-7B-Instruct, Quantization Bit=INT42025.0847.3
VLMQModel=Qwen2-VL-7B-Instruct, Bit-width=3-bit, Importance Factor Configuration=Default (ℓ1-norm)2025.0847.25
VLMQModel=LLaVA-OneVision-7B, Bit-width=3-bit, Importance Factor Configuration=token-level ℓ2-norm2025.0845.94
VLMQModel=LLaVA-OneVision-7B, Bit-width=3-bit, Importance Factor Configuration=Default (ℓ1-norm)2025.0845.41
GPTQModel=Qwen2-VL-7B-Instruct, Quantization Bit=INT42025.0844.03
GPTAQModel=Qwen2-VL-7B-Instruct, Bit-width=3-bit2025.0843.96
Full PrecModel=Qwen2-VL-2B-Instruct, Bit-width=Full Precision2025.0843.59
VLMQModel=Qwen2.5-VL-32B-Instruct, Bit-width=3-bit, Importance Factor Configuration=Default (ℓ1-norm)2025.0843.45
GPTAQModel=LLaVA-OneVision-7B, Bit-width=3-bit2025.0842.94
GPTAQModel=Qwen2.5-VL-32B-Instruct, Bit-width=3-bit2025.0842.67
GPTQModel=LLaVA-OneVision-7B, Bit-width=3-bit2025.0842.59
VLMQModel=Qwen2.5-VL-7B-Instruct, Bit-width=3-bit, Importance Factor Configuration=Default (ℓ1-norm)2025.0837.89
GPTAQModel=Qwen2.5-VL-7B-Instruct, Bit-width=3-bit2025.0836.49
VLMQBase Model=Qwen2-VL-7B-Instruct, Bit-width=2-bit, Group Size=128, Precursor Algorithm=GPTQ2025.0835.59
GPTQBase Model=Qwen2-VL-7B-Instruct, Bit-width=2-bit, Group Size=1282025.0834.11
GPTAQBase Model=Qwen2-VL-7B-Instruct, Bit-width=2-bit, Group Size=1282025.0833.8
VLMQModel=Qwen2-VL-2B-Instruct, Bit-width=3-bit, Importance Factor Configuration=Default (ℓ1-norm)2025.0832.58
GPTAQBase Model=LLaVA-OneVision-7B, Bit-width=2-bit, Group Size=1282025.0832.45
Full PrecModel=LLaVA-OneVision-0.5B, Bit-width=Full Precision2025.0832.13
VLMQBase Model=LLaVA-OneVision-7B, Bit-width=2-bit, Group Size=1282025.0831.38
GPTQBase Model=LLaVA-OneVision-7B, Bit-width=2-bit, Group Size=1282025.0831.35
GPTQModel=Qwen2-VL-2B-Instruct, Bit-width=3-bit2025.0830.83
VLMQBase Model=Qwen2.5-VL-7B-Instruct, Bit-width=2-bit, Group Size=128, Precursor Algorithm=GPTQ2025.0830.34
GPTAQModel=Qwen2-VL-2B-Instruct, Bit-width=3-bit2025.0828.71
GPTQModel=Qwen2-VL-7B-Instruct, Bit-width=3-bit2025.0827.9
GPTQModel=Qwen2.5-VL-7B-Instruct, Bit-width=3-bit2025.0827.33
GPTQModel=LLaVA-OneVision-0.5B, Bit-width=3-bit2025.0821.63
MBQModel=Qwen2-VL-7B-Instruct, Bit-width=3-bit2025.0821.35
AWQModel=Qwen2-VL-7B-Instruct, Bit-width=3-bit2025.0820.03
VLMQModel=LLaVA-OneVision-0.5B, Bit-width=3-bit, Importance Factor Configuration=‡ (unspecified variant)2025.0815.08
GPTQBase Model=Qwen2.5-VL-7B-Instruct, Bit-width=2-bit, Group Size=1282025.0813.89
GPTAQModel=LLaVA-OneVision-0.5B, Bit-width=3-bit2025.0811.27
AWQModel=Qwen2-VL-2B-Instruct, Bit-width=3-bit2025.088.48
MBQBase Model=Qwen2-VL-7B-Instruct, Bit-width=2-bit, Group Size=1282025.087.94
MBQModel=Qwen2-VL-2B-Instruct, Bit-width=3-bit2025.087.79
AWQBase Model=Qwen2-VL-7B-Instruct, Bit-width=2-bit, Group Size=1282025.087.47
GPTAQBase Model=Qwen2.5-VL-7B-Instruct, Bit-width=2-bit, Group Size=1282025.083.6