stable-diffusion text-to-image cubism

Cubism Diffusion

This is a meticulously fine-tuned version of the Stable Diffusion model, specifically built upon the foundations of its v2.1 iteration. Its unique training regimen revolves around the captivating realm of cubist arts, primarily drawing inspiration from the mesmerizing oeuvre of the renowned artist Pablo Picasso, with a predominant focus on his distinctive and groundbreaking paintings. To infuse your creative endeavors with the mesmerizing essence of this cubist aesthetic, you may seamlessly incorporate the specialized token cbst within your prompts at any desired juncture. By doing so, you can harness the transformative power of this model to craft artworks that encapsulate the essence and style reminiscent of cubism, as epitomized by Picasso himself. For instance, you can evoke this unique artistic genre in your requests, such as conjuring a "portrait of a human in the captivating cbst style."

Download the ckpt file from "files and versions" tab into the stable diffusion models folder of your web-ui of choice.

--

Faces rendered with this model: Face Samples

--

Cities and architectural works rendered with this model: Place Samples

--

Art pieces rendered with this model: Art Samples

--

This model was trained with Dreambooth, using ShivamShrirao Colab notebook named DreamBooth_Stable_Diffusion.ipynb

Usage

from diffusers import StableDiffusionPipeline
import torch

model_id = "bariscal/cbst_style"
pipe = StableDiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.float16)
pipe = pipe.to("cuda")

prompt = "portrait of human in cbst style"
image = pipe(prompt).images[0]

image.save("./human.png")

License

This model is freely accessible to everyone under the terms of the CreativeML OpenRAIL-M license, which defines specific rights and usage guidelines. The CreativeML OpenRAIL License entails the following:

  1. You are prohibited from using the model intentionally to produce or distribute illegal or harmful outputs or content.
  2. The authors make no claims to the outputs generated by the model; you are granted the freedom to use them, but you are responsible for their usage, which must adhere to the stipulations outlined in the license.
  3. You have the permission to redistribute the model's weights and utilize it for commercial purposes and/or as a service. However, if you choose to do so, please ensure that you incorporate the same usage restrictions as those specified in the license and provide a copy of the CreativeML OpenRAIL-M license to all your users. (Please review the entire license document thoroughly for complete details.) Please read the full license here