import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("cfu/sd2-turbo-bin", dtype=torch.bfloat16, device_map="cuda")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]
Stable Diffusion 2 Turbo stabilityai/sd-turbo, converted to .bin
There is a bug in my version of diffusers when converting from safetensors. So I convert the model into .bin and upload it here for the sake of convenience.
pipe, params = diffusers.pipelines.stable_diffusion.FlaxStableDiffusionPipeline.from_pretrained("sd-turbo",from_pt=True,use_safetensors=False)
- Downloads last month
- 5
Model tree for cfu/sd2-turbo-bin
Base model
stabilityai/sd-turbo