AI BriefingQwen-Image-2.1-Uncensored-Quantized: Run Qwen-Image 2.1 Locally with 4GB VRAM
abenzerps/Qwen-Image-2.1-Uncensored-Quantized
About the project
The original weights of Qwen-Image 2.1 are provided in various quantization formats such as GGUF and MLX. The safety filter has been removed, allowing direct generation of adult or sensitive content without prompt rejection.
The NVFP4 and MLX 4-bit versions have a file size of around 4GB, enabling lightweight execution on low-spec GPUs or Apple Silicon. By offloading the text encoder to system RAM via the ComfyUI-GGUF node, VRAM usage can be reduced to 9–17GB.
The repository includes the text encoder and VAE files, allowing immediate integration into workflows without separate downloads. The Q4_K_M format is recommended for the best balance of size and quality.
abenzerps/Qwen-Image-2.1-Uncensored-Quantized
The original page has no description.
text-to-image
This introduction was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report errors, attribution issues, or removal requests via Contact.