Skip to content
forked from mit-han-lab/hart

HART: Efficient Visual Generation with Hybrid Autoregressive Transformer

License

Notifications You must be signed in to change notification settings

vladmandic/hart

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

HART: Efficient Visual Generation with Hybrid Autoregressive Transformer

Credits: Paper | Project Page | Model

Install

./install.sh

Note

Install will create VENV and use UV to install all requirements
Models will be auto-downloaded on first use

Important

HART compute kernels need to be compiled and installed on first use
This requires CUDA and NVCC to be installed on the system

Run

source venv/bin/activate
python app.py

image

About

HART: Efficient Visual Generation with Hybrid Autoregressive Transformer

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 80.9%
  • Cuda 18.1%
  • Other 1.0%