Diffusers How to use JunkieMonkey69/Bella_ZimageTurbo with Diffusers:
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("JunkieMonkey69/Bella_ZimageTurbo")
prompt = "A dreamy Y2K-style night portrait of a young Caucasian woman, Bella. She poses playfully under a soft flash, her long voluminous hair cascading freely in loose waves over her shoulders. Her skin is naturally glowing and flawless. She is wearing a long, light blue crochet dress with thin straps, delicately adorned with 3D floral appliqués scattered throughout. A thin gold necklace and a single diamond ring complete the look, along with sleek earrings. Her nails are long, pointed, and French-manicured. Her gaze is slightly upward, lips gently smiling. One hand touches her hair while the other rests gracefully on her waist (or naturally repositioned for best composition). The photo captures the bold 90s aesthetic — camera flash highlighting her glowing skin, jewelry, and textured dress against a dark background. Makeup: nude tones with defined contours, long natural lashes, soft brown eyeliner, subtle blush, and matte nude lips outlined with a deep brown lip liner. Eyebrows are neatly styled with a Brow-lifting effect. Shot on Leica camera in ultra-realistic quality — hyper-detailed, cinematic lighting, sharp focus, Y2K-inspired atmosphere, shimmering yet elegant mood."
image = pipe(prompt).images[0] 
