List your company

Multimodal Understanding on Multimodal Evaluation Suite (MME, TextVQA, DocVQA, AI2D, MMMU, MMBench)

1,687.9MME ScoreQwen2.5-VL-7B
1,369.141,451.8951,534.651,617.405Jul 8, 2026

Evaluation results

MethodLinks
Qwen2.5-VL-7BVisual-token budget=1296 (100%), Base Model=Qwen2.5-VL-7B2026.071,687.977.394.383.250.683.180.5100
DivPruneVisual-token budget=256 (↓ 80.2%), Base Model=Qwen2.5-VL-7B, Venue=CVPR’252026.071,671.97067.977.848.68077.291.9
AnchorPruneVisual-token budget=128 (↓ 90.1%), Base Model=Qwen2.5-VL-7B2026.071,663.968.45177.746.879.675.788.1
AnchorPruneVisual-token budget=256 (↓ 80.2%), Base Model=Qwen2.5-VL-7B2026.071,657.572.672.978.848.680.878.293.5
DivPruneVisual-token budget=128 (↓ 90.1%), Base Model=Qwen2.5-VL-7B, Venue=CVPR’252026.071,642.466.349.875.247.376.975.486.6
FastVVisual-token budget=256 (↓ 80.2%), Base Model=Qwen2.5-VL-7B, Venue=ECCV’242026.071,638.974.96078.349.379.877.391.6
CDPrunerVisual-token budget=256 (↓ 80.2%), Base Model=Qwen2.5-VL-7B, Venue=NeurIPS’252026.071,621.963.252.77847.7787787.3
CDPrunerVisual-token budget=128 (↓ 90.1%), Base Model=Qwen2.5-VL-7B, Venue=NeurIPS’252026.071,572.157.736.376.945.977.974.782.3
AnchorPruneVisual-token budget=64 (↓ 95.1%), Base Model=Qwen2.5-VL-7B2026.071,563.261.531.374.245.275.773.280.8
FastVVisual-token budget=128 (↓ 90.1%), Base Model=Qwen2.5-VL-7B, Venue=ECCV’242026.071,554.870.74072.947.176.372.184
DivPruneVisual-token budget=64 (↓ 95.1%), Base Model=Qwen2.5-VL-7B, Venue=CVPR’252026.071,544.160.731.670.746.974.671.780
CDPrunerVisual-token budget=64 (↓ 95.1%), Base Model=Qwen2.5-VL-7B, Venue=NeurIPS’252026.071,431.751.622.873.644.773.972.976
FastVVisual-token budget=64 (↓ 95.1%), Base Model=Qwen2.5-VL-7B, Venue=ECCV’242026.071,381.463.727.768.445.467.564.775.3