Llamas on the Web: Memory-Efficient, Performance-Portable, and Multi-Precision LLM Inference with WebGPU

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Levine, Reese, Sharma, Rithik, Jain, Nikhil, Ramesh, Abhijit, Chen, Zheyuan, Abbas, Neha, Contini, James, Sorensen, Tyler
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!