ComfyUI-Z-Image-Turbo

Comfy UI nodes for Z-Image-Turbo to run native weights with automatic model downloading and support for 2048x2048 resolution. (Description by CC)

By tpc2233·Created 8 months ago·Updated 8 months ago· 14

tpc2233/ComfyUI-Z-Image-Turbo

Nodes2

On cloudLocal install

CategoryZ-Image-Turbo

Stars14

Updated8 months ago

Nodes (2)

Z-Image Turbo Loader (ModelScope)

The entry point for running native Z-Image weights

Z-Image-Turbo

Z-Image Turbo Sampler

One-shot generation from a native diffusers pipe

Z-Image-Turbo

Readme

<div align="center"> # ComfyUI-Z-Image-Turbo

Comfy UI nodes for Z-Image-Turbo to run native weights

</div>

✅ Install on Comfy via Manager: manger

✅ Install on Comfy manually:

cd custom_nodes
git clone https://github.com/tpc2233/ComfyUI-Z-Image-Turbo.git

cd ComfyUI-Z-Image-Turbo
pip install modelscope
pip install git+https://github.com/huggingface/diffusers
pip install -r requirements.txt

✅ Comfy Workflow:

https://github.com/tpc2233/ComfyUI-Z-Image-Turbo/blob/main/workflow.json

✅ Autodownload the models:

It will download the models first time you run workflow

It will be downloaded to: ComfyUI/models/diffusers/Z-Image-Turbo

✅ Run:

24Gb VRAM usage on 2048x2048

zturbo6