List your company
ResearchFormerly WizWand

Multi-modal Benchmark on MMBench Chinese (dev)

81.1AccuracySAIL-VL2
44.07653.68863.372.912May 20, 2024Aug 23, 2024Nov 27, 2024Mar 3, 2025Jun 7, 2025Sep 11, 2025Dec 16, 2025
Updated 8mo ago

Evaluation results

MethodLinks
SAIL-VL2Number of Parameters=2.7B2025.1281.1
HyperVL ViTLNumber of Parameters=2.0B2025.1279.64
HyperVLNumber of Parameters=1.8B2025.1279.2
InternVL3Number of Parameters=2.1B2025.1278.4
AndesVLNumber of Parameters=2.4B2025.1277.8
Qwen3-VLNumber of Parameters=2.1B2025.1276.28
Qwen2-VLNumber of Parameters=2.2B2025.1273.5
Ovis2Number of Parameters=2.5B2025.1272.2
SAIL-VL1.5Number of Parameters=2.5B2025.1271.7
Yi-VL-6BLLM=Yi-6B, Visual Encoder=CLIP-H@224, #PT=100M, #FT=26M2024.0568.6
MiniCPM-V-3BLLM=MiniCPM-SFT-2B, Visual Encoder=SigLIP-SO@384, #PT=300M, #FT=8M2024.0565.3
LLaVA-1.5-13BLLM=Vicuna-1.5-13B, Visual Encoder=CLIP-L@336, #PT=558K, #FT=665K2024.0561.5
Imp-2BLLM=Qwen-1.5-1.8B, Visual Encoder=SigLIP-SO@384, #PT=558K, #FT=1M2024.0561.3
Imp-4BLLM=Phi-3-3.8B, Visual Encoder=SigLIP-SO@384, #PT=558K, #FT=1M2024.0561.1
LLaVA-1.5-7BLLM=Vicuna-1.5-7B, Visual Encoder=CLIP-L@336, #PT=558K, #FT=665K2024.0558.9
Bunny-2BLLM=Qwen1.5-1.8B, Visual Encoder=SigLIP-SO@384, #PT=2M, #FT=695K2024.0558.5
Imp-3BLLM=Phi-2-2.7B, Visual Encoder=SigLIP-SO@384, #PT=558K, #FT=1M2024.0546.7
InternVL3.5Number of Parameters=2.3B2025.1245.5