How to use from the
Use from the
Diffusers library
pip install -U diffusers transformers accelerate
import torch
from diffusers import DiffusionPipeline

# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("Tongyi-MAI/Z-Image-Turbo", dtype=torch.bfloat16, device_map="cuda")
pipe.load_lora_weights("Shakker-Labs/AWPortrait-Z")

prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]

AWPortrait-Z

AWPortrait-Z is a portrait-beauty LoRA meticulously built on the Z-Image.

  • Native-noise reduction: fixed Zimage’s chronic grain—those downy, high-frequency artifacts that plagued skin tones—so complexions now look flawlessly real.
  • Relit lighting: tamed the base model’s excessive HDR, restoring punchy contrast and saturation; re-engineered artificial-light behavior so studio strobes sit naturally in-scene instead of floating above it.
  • Diverse faces: expanded multi-ethnic feature coverage, breaking the “same-face” barrier and delivering portraits that are both authentic and unmistakably individual.

Showcases

Comparison

Acknowledgements

This model is released by DynamicWang.

Downloads last month
885
Inference Providers NEW

Model tree for Shakker-Labs/AWPortrait-Z

Adapter
(667)
this model
Adapters
7 models

Spaces using Shakker-Labs/AWPortrait-Z 13