Ragged Paged Attention: A High-Performance and Flexible LLM Inference Kernel for TPU

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Jiang, Jevin, Chen, Ying, Hechtman, Blake A., Zhang, Fenghui, Mu, Yarong
Format:	Preprint
Published:	2026
Subjects:	Performance Artificial Intelligence Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Internet

https://arxiv.org/abs/2604.15464