List your company
ResearchFormerly WizWand

Visual Perception on MMBench (T)

79.1Accuracy (MMBench T)MG-LLaVA
30.32442.98755.6568.313Jun 25, 2024
Updated 8mo ago

Evaluation results

MethodLinks
MG-LLaVALLM=Yi1.5-34B, Params.=34.4B, Res.=336 (768)2024.0679.1
GPT-4V2024.0677
MG-LLaVALLM=LLaMA3-8B, Params.=8.4B, Res.=336 (768)2024.0676.6
GeminiPro Vision2024.0673.6
MG-LLaVALLM=Vicuna-13B, Params.=13.6B, Res.=336 (768)2024.0673.5
MG-LLaVALLM=Phi3-3.8B, Params.=4.2B, Res.=336 (768)2024.0672.9
MG-LLaVALLM=Vicuna-7B, Params.=7.4B, Res.=336 (768)2024.0671.9
LLaVA-1.5LLM=Vicuna-13B, Params.=13.4B, Res.=3362024.0669.2
MiniCPM-V2LLM=MiniCPM-2.4B, Params.=2.8B, Res.=4482024.0669.1
Qwen-VL-Plus2024.0667
LLaVA-1.5LLM=Vicuna-7B, Params.=7.2B, Res.=3362024.0666.5
Qwen-VL-ChatLLM=Qwen-7B, Params.=9.6B, Res.=4482024.0661.8
ShikraLLM=Vicuna-13B, Params.=7.3B, Res.=2242024.0660.2
IDEFICS-9BLLM=LLaMA-7B, Res.=2242024.0645.3
Qwen-VLLLM=Qwen-7B, Params.=9.6B, Res.=4482024.0632.2