Skip to content

This is the official PyTorch implementation of "ZipAR: Accelerating Auto-regressive Image Generation through Spatial Locality"

Notifications You must be signed in to change notification settings

ThisisBillhe/ZipAR

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 

Repository files navigation

ZipAR: Accelerating Auto-regressive Image Generation through Spatial Locality

This repository provides the implementation for our paper "ZipAR: Accelerating Auto-regressive Image Generation through Spatial Locality". Our approach can greatly accelerate the generation speed of auto-regressive (AR) visual generation models.

arXiv

Getting Started

Install the pre-trained AR visual generation models and ZipAR only modifies several lines of the code.

BibTeX

If you find this work useful for your research, please consider citing:

@article{he2024zipar,
  title={ZipAR: Accelerating Auto-regressive Image Generation through Spatial Locality},
  author={He, Yefei and Chen, Feng and He, Yuanyu and He, Shaoxuan and Zhou, Hong and Zhang, Kaipeng and Zhuang, Bohan},
  journal={arXiv preprint arXiv:2412.04062},
  year={2024}
}

About

This is the official PyTorch implementation of "ZipAR: Accelerating Auto-regressive Image Generation through Spatial Locality"

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published