hugpadhugpad

z-lab / Qwen3-4B-DFlash-b16

Text Generation537M19.5K31Updated Apr 7, 2026
Hugging Face

Downloads

19.5K

Likes

31

Last updated

Apr 7, 2026

License

mit

DFlash is a novel speculative decoding method that utilizes a lightweight block diffusion model for drafting. It enables efficient, high-quality parallel drafting that pushes the limits of inference speed.

Tasks & library

  • text generation
  • transformers

Tags

  • transformers
  • qwen3
  • feature-extraction
  • dflash
  • speculative-decoding
  • diffusion

Technical specifications

Parameters537M
Librarytransformers
LanguagesNot available
Licensemit