Lynx: Enabling Efficient MoE Inference through Dynamic Batch-Aware
Expert Selection

Lynx: Enabling Efficient MoE Inference through Dynamic Batch-Aware Expert Selection

13 November 2024

Ada Gavrilovska

Anand Padmanabha Iyer

ArXiv (abs)PDF HTML

Papers citing "Lynx: Enabling Efficient MoE Inference through Dynamic Batch-Aware Expert Selection"

Title
No papers