HermiHg author hub
Qwen3.8 27B DFlash2 — Q2 K S MIX draft half the reference size A mixed precision 2–3 bit quantization of the DFlash 2 draft model for Qwen/Qwen3.8 27B https://huggingface.co/Qwen/Qwen3.8 27B , built to be 50% the size of the reference Q4 K M checkpoint while retaining ~98% of it…
Models
2
Downloads
5,069
Run models locally with guIDE
Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.