Uncensored vision-enabled Qwen3.6-27B finetune trends on Hugging Face with GGUF release

HauhauCS · hf · 2026-08-16

Community user HauhauCS has released an uncensored finetune of Qwen/Qwen3.6-27B, "Qwen3.6-27B-Uncensored-HauhauCS-Aggressive", now trending on Hugging Face.

The model is an image-text-to-text pipeline, accepting both image and text input, and ships in GGUF format for easy local deployment with llama.cpp/Ollama. It supports Chinese, English and other languages. The "Aggressive" in its name signals an aggressively guardrail-stripped variant — a common pattern among community uncensored derivatives, with unverified quality.

Original post →

More from Models

Models channel →