List your company
ResearchFormerly WizWand

Open-ended generation on MM-Vet

45.55MM-Vet ScoreBaseline
22.846828.740934.63540.5291May 19, 2026
Updated 5mo ago

Evaluation results

MethodLinks
BaselineBase Model=QWEN2.5-VL-7B-INSTRUCT2026.0545.55
VisionZip + RotateKBase Model=QWEN2.5-VL-7B-INSTRUCT, Token Sparsity Ratio=0.35, Channel Sparsity Ratio=0.252026.0544.04
FastV + RotateKBase Model=QWEN2.5-VL-7B-INSTRUCT, Token Sparsity Ratio=0.30, Channel Sparsity Ratio=0.252026.0541.24
VisionZip + ThinKBase Model=QWEN2.5-VL-7B-INSTRUCT, Token Sparsity Ratio=0.35, Channel Sparsity Ratio=0.252026.0535
VisionZipBase Model=QWEN2.5-VL-7B-INSTRUCT, Token Sparsity Ratio=0.222026.0534.77
FastVBase Model=QWEN2.5-VL-7B-INSTRUCT, Token Sparsity Ratio=0.18752026.0532.61
FastV + ThinKBase Model=QWEN2.5-VL-7B-INSTRUCT, Token Sparsity Ratio=0.30, Channel Sparsity Ratio=0.252026.0532.61
BaselineBase Model=LLAMA3-LLAVA-NEXT-8B2026.0527.14
FastV + RotateKBase Model=LLAMA3-LLAVA-NEXT-8B, Token Sparsity Ratio=0.30, Channel Sparsity Ratio=0.252026.0526.97
VisionZip + RotateKBase Model=LLAMA3-LLAVA-NEXT-8B, Token Sparsity Ratio=0.35, Channel Sparsity Ratio=0.252026.0526.97
FastV + ThinKBase Model=LLAMA3-LLAVA-NEXT-8B, Token Sparsity Ratio=0.30, Channel Sparsity Ratio=0.252026.0525.05
VisionZipBase Model=LLAMA3-LLAVA-NEXT-8B, Token Sparsity Ratio=0.222026.0524.59
FastVBase Model=LLAMA3-LLAVA-NEXT-8B, Token Sparsity Ratio=0.18752026.0524.5
VisionZip + ThinKBase Model=LLAMA3-LLAVA-NEXT-8B, Token Sparsity Ratio=0.35, Channel Sparsity Ratio=0.252026.0523.72