import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("ShadoWxShinigamI/Midjourney-Shatter", dtype=torch.bfloat16, device_map="cuda")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]##MODEL BY ShadoWxShinigamI
Use Token - mdjrny-shttr at the beginning of your prompt [ Prompt Engineering not required ]; If some object doesn't work, increase the prompt weight of the object to 1.6
Training - 2500 steps, Batch size 2, 512x512, v1-5 Base, 26 images (52 Flipped)
Examples:-
Lion
Batman
Medusa Head
Deer
Emma Watson
Son Goku
(Mansion:1.6) - Need the extra weight for the model to actually produce buildings for some reason
- Downloads last month
- 48