Skip to content

EnderXie23/D2C

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

D2C: Dual Dynamic Cache for dLLM optimization

This is the course project repo for AI+X: Speeding up LLMs, Summer 2025, Tsinghua University.

D2C is a novel approach to optimize diffusion Large Language Models (dLLMs) by leveraging dual dynamic caching mechanisms. This repository contains the implementation of D2C, along with instructions on how to use it.

See our course report here.

About

D2C: Dual Dynamic Cache for dLLM optimization

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors