brushpenbob / splitstyle-2-0

Splitstyle 2.0

Model description

Second attempt at creating a LoRa that recruits for different traditional art styles.

I would recommend having the CFG scale set pretty precise, and the step counts relatively high

I realized after the training was already completed. I trained us as a style and not a concept so this is the second attempt added a few more images to the data set up and that helps.

It should now allow you the option to create either two panels or four panel images. with the triggers " 2splitstyle" or "4splitstyle"

Trained on close-up portraits. Might take a bit more tinkering as it tends to just have two panels instead of four. anyway while i work out the bugs enjoy

Trigger words

You should use 2splitstyle , 4splitstyle , splitstyle , split_style , evang to trigger the image generation.

Download model

Weights for this model are available in Safetensors format.

Download them in the Files & versions tab.

Use it with the 🧨 diffusers library

from diffusers import AutoPipelineForText2Image
import torch

pipeline = AutoPipelineForText2Image.from_pretrained('stabilityai/stable-diffusion-xl-base-1.0', torch_dtype=torch.float16).to('cuda')
pipeline.load_lora_weights('brushpenbob/splitstyle-2-0', weight_name='Splitstyle_2.0.safetensors')
image = pipeline('`2splitstyle`, `4splitstyle`, `splitstyle`, `split_style`, `evang`').images[0]

For more details, including weighting, merging and fusing LoRAs, check the documentation on loading LoRAs in diffusers