---
license: mit
base_model:
- z-lab/Qwen3.6-27B-DFlash
tags:
  - dflash
  - speculative-decoding
  - diffusion
  - efficiency
  - flash-decoding
  - qwen
  - diffusion-language-model
---
This repository stores Z Lab's Qwen3.6 27B DFlash drafter quantized to Q4_K_M and Q8_0 using upstream llama.cpp and following [spiritbuun's guide](https://huggingface.co/spiritbuun/Qwen3.6-27B-DFlash-GGUF).