Qwen3.8-27B-DFlash2-mxfp4

Description

Qwen3.8-27B-DFlash2-mxfp4 is an open-weight text-generation model with roughly 27B parameters published by deresolution on the Hugging Face Hub. It has 785 downloads. Runs with mlx.

Categories

• Coding

Tags

• qwen3
• apple-silicon
• mxfp4
• quantized
• dflash2
• speculative-decoding
• block-diffusion
• draft-model

Capabilities

• Text generation
• Library: mlx
• 27B parameters