Batch-Max: Higher LLM Throughput using Larger Batch Sizes and KV Cache Compression

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Metel, Michael R., Chen, Boxing, Rezagholizadeh, Mehdi
Format:	Preprint
Published:	2024
Subjects:	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/2412.05693