import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("calcuis/illustrious", dtype=torch.bfloat16, device_map="cuda")
pipe.load_lora_weights("Ava2000/Gravitydusty_Illustrious")
prompt = "masterpiece, best quality, shiny skin, oil painting \\(medium\\),"
image = pipe(prompt).images[0]GravityDusty Style - Illustrious
- Prompt
- masterpiece, best quality, shiny skin, oil painting \(medium\),
- Negative Prompt
- text, long body, lowres, cropped, lowres, jpeg artifacts, (signature, artist name, watermark, username, english text, signature:1.4), simple background, conjoined, out of frame, out of frame, censored, bar censor,
- Prompt
- masterpiece, best quality, shiny skin, oil painting \(medium\),
- Negative Prompt
- text, long body, lowres, cropped, lowres, jpeg artifacts, (signature, artist name, watermark, username, english text, signature:1.4), simple background, conjoined, out of frame, out of frame, censored, bar censor,
Model description
I couldn't get the the signatures and watermarks out of the dataset so these are kinda baked in. Using the following prompts in the negative can help: signature, watermark, artist name,
Trigger: gravitydusty style,
Extra triggers for style: painting (medium), watercolor (medium), shiny skin, oil painting (medium),
Advised strength: 0.6 - 1.0
Trigger words
You should use gravitydusty style to trigger the image generation.
Download model
Download them in the Files & versions tab.
- Downloads last month
- 8
Model tree for Ava2000/Gravitydusty_Illustrious
Base model
KBlueLeaf/kohaku-xl-beta5 Quantized
calcuis/illustrious