ExpertFlow: Efficient Mixture-of-Experts Inference via Predictive Expert Caching and Token Scheduling

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	He, Xin, Zhang, Shunkang, Tang, Kaijie, Shi, Shaohuai, Wang, Yuxin, Zeng, Zihao, Tang, Zhenheng, Chu, Xiaowen, Yin, Haiyan, Tsang, Ivor W., Ong, Yew Soon
Format:	Preprint
Published:	2024
Subjects:	Artificial Intelligence Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Internet

https://arxiv.org/abs/2410.17954