SDArt : Complete Edition (version based on 1.5)

Model description

This is a model related to the community events on Stable Diffusion discord.

Each week multiple contests happen and challengers submit great art that I train on as individual models. The present model is a compilation of all those datasets, as well as some collected through a community project.

The total dataset is made of 742 pictures. It was trained on Stable diffusion 1.5 with the updated VAE. I used EveryDream to do the training, 100 total repeat per picture. The pictures were tagged using long prompts, and no specific trigger word is needed. The name of the contests themselves have been added to the caption and can act as styles (Synesthesia, Futuristic clothes, ...)

This model is also available here in a version trained on 2.1 as a base.

Trained themes

Alice In Diffusion Land - individual model 1.5 2.1 => Example results of the complete model
Elemental Flowers - individual model 1.5 => Example results of the complete model
Synesthesia - individual model 1.5 2.1 => Example results of the complete model
Chaos & Order - individual model 1.5 2.1 => Example results of the complete model
Futuristic Fashion - individual model 1.5 => Example results of the complete model
Cosmic horrors - individual model 1.5 2.1 => Example results of the complete model
Encapsulated - individual model 1.5 2.1 => Example results of the complete model
Unmythical Creatures - individual model 1.5 2.1 => Example results of the complete model
Catstravaganza - individual model 1.5 => Example results of the complete model
Someone - individual model 1.5 => Example results of the complete model
Something - individual model 1.5 => Example results of the complete model
Somewhere - individual model 1.5 => Example results of the complete model
Underwater Worlds - individual model 1.5 => Example results of the complete model
Abstract Realities - individual model 1.5 => Example results of the complete model
Imaginary Friend - individual model 1.5 => Example results of the complete model
The Never Ending Loop - individual model 1.5 => Example results of the complete model
as well as around 60 favorite from the community => Example results of the complete model

Download links

SafeTensors

Dataset

🧨 Diffusers

This model can be used just like any other Stable Diffusion model. For more information, please have a look at the Stable Diffusion.

You can also export the model to ONNX, MPS and/or FLAX/JAX.

from diffusers import StableDiffusionPipeline
import torch

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

prompt = "a working cat, wearing a suit, using a computer, technological background, tea cup, painting, Catstravaganza"
image = pipe(prompt).images[0]

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