List your company

Prefill Latency on MMMU

1,797.4Prefill Latency (ms)Pruning Only
1,757.4362,027.1932,296.952,566.707Nov 22, 2025

Evaluation results

MethodLinks
Pruning OnlyRetain=25%, K_v (Experts)=8, Batch Size=8, Hardware=NVIDIA A800 GPU, Backbone=InternVL3.52025.111,797.41.56
FastMMoERetain=75%, K_v (Experts)=4, Batch Size=8, Hardware=NVIDIA A800 GPU, Backbone=InternVL3.52025.111,800.31.55
Pruning OnlyRetain=50%, K_v (Experts)=8, Batch Size=8, Hardware=NVIDIA A800 GPU, Backbone=InternVL3.52025.111,817.41.54
Pruning OnlyRetain=75%, K_v (Experts)=8, Batch Size=8, Hardware=NVIDIA A800 GPU, Backbone=InternVL3.52025.111,827.21.53
BaselineRetain=100%, K_v (Experts)=8 (Full), Batch Size=8, Hardware=NVIDIA A800 GPU, Backbone=InternVL3.52025.112,796.51