artwork: testing the Z-IMAGE model
category: model check
base models: Z-IMAGE-Turbo by Tongyi-MAI
workflow: powered by ComfyUI
influenced: pure versus trained LoRAs mit own photography
Z-Image: An Efficient Image Generation Foundation Model with Single-Stream Diffusion Transformer
Model generates 2024×2048 px in a time that is easy to wait for!
More about the model:
HUGGING FACE
https://huggingface.co/Tongyi-MAI/Z-Image-Turbo
„A distilled version of Z-Image that matches or exceeds leading competitors with only 8 NFEs (Number of Function Evaluations). It offers sub-second inference latency on enterprise-grade H800 GPUs and fits comfortably within 16G VRAM consumer devices. It excels in photorealistic image generation, bilingual text rendering (English & Chinese), and robust instruction adherence.“
License: Apache License Version 2.0
https://huggingface.co/datasets/choosealicense/licenses/blob/main/markdown/apache-2.0.md
Non-edited images:
. . . first generation using ComfyUI Workflow with borrowed prompt for the cover description . . . 4 steps, cfg 1, sampler res_multistep, scheduler simple, denoise 1 . . . 2048×2048 px . . .
. . . 8 steps, cfg 1, sampler euler, scheduler simple, denoise 1 . . . guess this is the reason for the brighter colors . . .
. . . 8 steps, cfg 1, sampler euler, scheduler simple, denoise 1 . . .
.