superdiff/superdiff-sdxl-v1-0

The Superposition of Diffusion Models Using the Itô Density Estimator: Pipeline

This pipeline shows how to superimpose different text prompts from Stable Diffusion-XL 1.0 based the paper The Superposition of Diffusion Models Using the Itô Density Estimator. The authors would like to thank Viktor Ohanesian for developing the SD-XL pipeline.

Requirements

This pipeline can be run with the following packages & versions:

• PyTorch 2.5.1
• Diffusers 0.32.1
• Accelerate 1.2.1
• Transformers 4.47.1

You can install these with:

pip install torch
pip install diffusers accelerate transformers

Example usage

from PIL import Image
from diffusers import DiffusionPipeline

pipeline = DiffusionPipeline.from_pretrained("superdiff/superdiff-sdxl-v1-0", custom_pipeline='pipeline', trust_remote_code=True)
output = pipeline("a flamingo", "a candy cane", seed=1, num_inference_steps=200, batch_size=1)

image = Image.fromarray(output[0])
image.save("superdiff_output.png")

Arguments that can be set by user in pipeline():

• prompt_1 [required]: text prompt describing first concept to superimpose (e.g. "a flamingo")
• prompt_2[required]: text prompt describing second concept to superimpose (e.g. "a candy cane")
• seed[optional: default=None]: seed for random noise generator for reproducibility; for non-deterministic outputs, set to None
• num_inference_steps[optional: default=200]: number of denoising steps
• batch_size [optional: default=1]: batch size
• guidance_scale [optional: default=7.5]: scale for classifier-free guidance
• height, width [optional: default=1024]: height and width of generated images (we recommend leaving it at 1024!)

Note: for generating realistic photos with SDXL, we recommend using prompts such as "teapot, high quality photography" or "a highly realistic photo of a volcano".

Citation

BibTeX:

@article{skreta2025superposition,
  title={The Superposition of Diffusion Models Using the It$\backslash$\^{} o Density Estimator},
  author={Skreta, Marta and Atanackovic, Lazar and Bose, Avishek Joey and Tong, Alexander and Neklyudov, Kirill},
  journal={International Conference on Learning Representations},
  year={2025}
}