abenzerps/Qwen-Image-2.1-Uncensored-GGUF — new model trending #30 on Hugging Face
An uncensored GGUF version of Qwen-Image-2.1 for local image generation is trending, featuring no content filters and optimized for ComfyUI.
A trending Hugging Face repository provides GGUF quantizations of the Qwen-Image-2.1 model for local image generation using ComfyUI. The release is explicitly 'uncensored,' lacking built-in safety checkers or content filters, and will generate sensitive imagery based on prompts. It includes various quantization levels (Q4_K_M recommended) and companion files, with a note about a current bug affecting the Q8_0 version.
- New GGUF quantizations of Qwen-Image-2.1 for local image generation are trending on Hugging Face.
- The release is an 'uncensored' version without safety filters or content restrictions.
- Quantizations range from Q4_0 (4.05 GB) to Q8_0 (7.59 GB) for use with ComfyUI.
- A known issue with Q8_0 quantization causes tensor shape mismatches in ComfyUI sampling.
- The model requires companion text encoder and VAE files for operation.
- Vendors
- Qwen
- Products
- Qwen-Image-2.1-Uncensored-GGUF
- Organizations
- Hugging FaceComfy-OrgComfyUI
- AI models
- Qwen-Image-2.1
From the original article
3 min read at huggingface.co> Running `qwen-image-2. 1-Q8_0. gguf` in ComfyUI currently triggers a tensor shape mismatch error (`[136] vs [128]`) during sampling due to 1D RMSNorm scale quantization.
By abenzerps · huggingface.co
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.