import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-xl-base-1.0", dtype=torch.bfloat16, device_map="cuda")
pipe.load_lora_weights("alvdansen/japanese-poster-prints")
prompt = "bubble tea, minimalist, daiton style, block print"
image = pipe(prompt).images[0]Japenese Poster Prints
- Prompt
- bubble tea, minimalist, daiton style, block print
- Negative Prompt
- signature, watermark, kanji
- Prompt
- a bird, daiton style
- Prompt
- a swallow, daiton style
- Prompt
- a toad
- Prompt
- a happy kid, minimalist, daiton style, block print
- Negative Prompt
- signature, watermark, kanji
Model description
This model does a mid century Japanese block print style. If you have a more complex subject, you can use the trigger "daiton" to make the style stronger.
If you are seeing unusual characters or text, just use neg prompt: signature, watermark, kanji
Trigger words
You should use daiton style to trigger the image generation.
You should use block print to trigger the image generation.
Download model
Weights for this model are available in Safetensors format.
Model release is for research purposes only. For commercial use, please contact me directly.
Download them in the Files & versions tab.
- Downloads last month
- 14
Model tree for alvdansen/japanese-poster-prints
Base model
stabilityai/stable-diffusion-xl-base-1.0