imgGen.py

import torch
from diffusers import StableDiffusionXLPipeline, UNet2DConditionModel, EulerDiscreteScheduler
from huggingface_hub import hf_hub_download
from safetensors.torch import load_file


base = "stabilityai/stable-diffusion-xl-base-1.0"
repo = "ByteDance/SDXL-Lightning"
ckpt = "sdxl_lightning_4step_unet.safetensors" # Use the correct ckpt for your step setting!


# Load model.
unet = UNet2DConditionModel.from_config(base, subfolder="unet").to("cuda", torch.float16)
unet.load_state_dict(load_file(hf_hub_download(repo, ckpt), device="cuda"))
pipe = StableDiffusionXLPipeline.from_pretrained(base, unet=unet, torch_dtype=torch.float16, variant="fp16").to("cuda")

# Ensure sampler uses "trailing" timesteps.
pipe.scheduler = EulerDiscreteScheduler.from_config(pipe.scheduler.config, timestep_spacing="trailing")

def generateTransparentImage(text):
    # Ensure using the same inference steps as the loaded model and CFG set to 0.
    image = pipe(text+', full length, one, transparent background, vibrant', num_inference_steps=4, guidance_scale=0).images[0]
    return image

if __name__ == "__main__":
    text = "a cat"
    img = generateTransparentImage(text)
    img.save("output.png")