metadata

license: cc-by-nc-sa-4.0
tags:
  - common-canvas
  - stable-diffusion
  - sdxl
datasets:
  - common-canvas/commoncatalog-cc-by-sa
  - common-canvas/commoncatalog-cc-by
  - common-canvas/commoncatalog-cc-by-nc-sa
  - common-canvas/commoncatalog-cc-by-nc
language:
  - en

CommonCanvas-XL-NC 0.1

Specifications

Input: CommonCatalog Text Captions
Output: CommonCatalog Images
Architecture: Stable Diffusion XL
Credit: CommonCanvas, StabilityAI, mosaicML, @multimodalart, @Wauplin, @lhoestq
NSFW: Yes
Text: https://arxiv.org/abs/2310.16825
LICENSE:

This work is licensed under CC BY-NC-SA 4.0 creative commons license logo terms of license logo non-commercial use logo share alike logo

Details

training data : Flickr100M dataset
bias : internet connected Western countries
limitations : text generation, complex composition, faces, non-English languages, VAE
use : research, deployment, examination, art, education, creative use
prohibited : commercial use
suggested training : mosaicML https://github.com/mosaicml/diffusion.
Citation

@article{gokaslan2023commoncanvas,
  title={CommonCanvas: An Open Diffusion Model Trained with Creative-Commons Images},
  author={Gokaslan, Aaron and Cooper, A Feder and Collins, Jasmine and Seguin, Landan and Jacobson, Austin and Patel, Mihir and Frankle, Jonathan and Stephenson, Cory and Kuleshov, Volodymyr},
  journal={arXiv preprint arXiv:2310.16825},
  year={2023}
}

Code

from diffusers import StableDiffusionXLPipeline
pipe = StableDiffusionXLPipeline.from_pretrained(
    "common-canvas/CommonCanvas-XL-NC", 
    custom_pipeline="multimodalart/sdxl_perturbed_attention_guidance", #read more at https://huggingface.co/multimodalart/sdxl_perturbed_attention_guidance
    torch_dtype=torch.float16
).to(device)

prompt = "a cat sitting in a car seat"
image = pipe(prompt, num_inference_steps=25).images[0]

CommonCanvas-XL-NC 0.1

Specifications

Details

Citation

Code