This PR make some changes to the config to be able to load the pipeline directly from diffusers.

from diffusers import StableDiffusionXLPipeline

device = "cuda"
model_id = "playgroundai/playground-v2.5-1024px-aesthetic"

pipe = StableDiffusionXLPipeline.from_pretrained(
    model_id,  torch_dtype=torch.float16,   variant="fp16",
).to(device)

images = pipe("...", num_inference_steps=25, guidance_scale=5, generator=generator).images
valhalla changed pull request status to open
Playground org

thanks a lot!

ehsanakh changed pull request status to merged

Sign up or log in to comment