potatopizza/Stable-Diffusion-v1-4-wikiart

Model Card: Stable Diffusion v1-4 Fine-tuned on WikiArt

This model is a fine-tuned version of the Stable Diffusion v1.4 model on the WikiArt dataset.
It can generate images in various artistic styles learned from WikiArt.

Model Details

Model Description

• Developed by: potatopizza
• Funded by [optional]: N/A
• Shared by [optional]: potatopizza
• Model type: Text-to-Image
• License: CreativeML Open RAIL-M
• Finetuned from model: CompVis/stable-diffusion-v1-4

Model Sources

• Repository (Hugging Face): potatopizza/Stable-Diffusion-v1-4-wikiart

Uses

Direct Use

• Text-to-Image generation, specifically in various WikiArt styles.

Out-of-Scope Use

• Generating harmful, explicit, or disallowed content per the license.
• Any usage that violates the CreativeML Open RAIL license terms.

Bias, Risks, and Limitations

• The generated images might reflect biases present in the WikiArt dataset (art style bias, cultural bias, etc.).
• Potential misuse for generating disallowed content

Example Inference Result

Below is an example image generated using the prompt: A painting of the Eiffel Tower in the style of Eric Fischl.

How to Get Started with the Model

Use Model code:

from diffusers import StableDiffusionPipeline
import torch

model_id = "potatopizza/Stable-Diffusion-v1-4-wikiart"  
pipe = StableDiffusionPipeline.from_pretrained(
    model_id,
    torch_dtype=torch.float16
).to("cuda")

prompt = "A beautiful painting of a sunset over the mountains, in the style of eric fischl"
image = pipe(prompt).images[0]
image.save("example.png")