From https://modelscope.cn/models/nunchaku-tech/nunchaku-z-image-turbo/summary

This repository contains the Nunchaku quantized version of Z-Image-Turbo, a high-performance image generation model. This quantized version is optimized for efficient inference while maintaining minimal performance loss.

No news yet. Stay tuned for future updates!

Model Details Model Description Developed by: Nunchaku Team (Thanks @devgdovg) Model Type: Image-to-image License: apache-2.0 Base Model for Quantization: Z-Image-Turbo Model Files Data Type: INT4 for non-Blackwell GPUs (before 50-series), NVFP4 for Blackwell GPUs (50-series). Rank:

r32 for faster inference, r128 for better quality with slower inference, r256 for the highest quality (slowest inference).