Diffusers How to use cptsl/2000Retro 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("cptsl/2000Retro")
prompt = " A photograph captures a young woman with reddish-brown, shoulder-length wavy hair as the central subject. She is positioned in the foreground, facing the camera with her head slightly turned to her right. Her facial features include light-colored eyes, defined eyebrows, and lips parted as if mid-speech. She wears a gray zip-up hoodie over a t-shirt with a partially visible graphic. Her right arm is extended out of frame, and her right hand is blurred in motion, suggesting gesturing. Her body is angled slightly away from the viewer, with her left shoulder more prominent. The setting is an indoor hallway, likely a school, characterized by a light-colored cinder block wall to her right. This wall features a large corkboard covered with various flyers and notices in different colors and sizes, including orange, yellow, and white paper. To her left, the hallway extends into the background, populated by several other students who are out of focus. One person in a black shirt and blue jeans with a backpack is visible on the far left. The floor is a reflective, polished surface. The ceiling is composed of white tiles with rectangular fluorescent light fixtures. The image has a date stamp in red text at the bottom right corner: \"OCT 12 2001\". The image has a noticeable grain and a shallow depth of field, rendering the background and the woman's hand blurred while keeping her face relatively sharp. The aspect ratio is approximately 1:1."
image = pipe(prompt).images[0] 
