import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("sr5434/SDXL-v1.0-sfx-step-800", dtype=torch.bfloat16, device_map="cuda")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]SDXL 1.0 finetunes on vucinatim/spectrogram-captions for 89 epochs(800 steps). It generates spectrograms for simple sounds. It currently does not produce very good sound effects, but I will train the model for longer in the future.
- Downloads last month
- 4