This is an <a target="_blank" rel="ugc" href="https://huggingface.co/huihui-ai/Huihui-Qwen3-4B-abliterated-v2">Huihui uncensored version</a> of Qwen/Qwen3-4B created with abliteration (see <a target="_blank" rel="ugc" href="https://github.com/Sumandora/remove-refusals-with-transformers">remove-refusals-with-transformers</a> to know more about it). This is a crude, proof-of-concept implementation to remove refusals from an LLM model without using TransformerLens. Use this WorkFlow with it for maximum details: <a target="_blank" rel="ugc" href="https://civarchive.com/models/2199464">https://civarchive.com/models/2199464</a> Update: NSFW Z-image&gt; <a target="_blank" rel="ugc" href="https://huggingface.co/tewea/z_image_turbo_bf16_nsfw/tree/main">https://huggingface.co/tewea/z_image_turbo_bf16_nsfw/tree/main</a>

Z-Image Uncensored Text Encoder - Abliterated Huihui Qwen3 4B v2 (Q_8 GGUF)

read more: <a target="_blank" rel="ugc" href="https://huggingface.co/huihui-ai/Huihui-Qwen3-4B-abliterated-v2">https://huggingface.co/huihui-ai/Huihui-Qwen3-4B-abliterated-v2</a>

Files with SHA256: e0c5bafc6e7d699d499f4bfd96745b63b0492559397bbda8578715934d053f87