import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("michaelevanssavitar/Physics", torch_dtype=torch.bfloat16, device_map="cuda")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]Physics: Wan Factorized OPD
This repository packages Wan2.1-T2V-1.3B, the frozen v3 Appearance/Dynamic factorizer, WISA prompt metadata, and the v6.1 corrected-SDE routed OPD training code.
Use revision v6.1.1-balanced-query and follow
physics/OPD_TRAINING_GUIDE.md.
The default OPD launcher starts the student from base Wan. The legacy
step-000400.pt checkpoint is included for explicit ablations only and is not
loaded by default.
- Downloads last month
- -
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support