Efficient-Large-Model

/

Sana_1600M_1024px

like 151

Follow

Efficient-Large-Model 201

I guess use diffusers when it is ready?

import spaces
from diffusers import DiffusionPipeline

pipe = DiffusionPipeline.from_pretrained(Efficient-Large-Model/Sana_1600M_1024px_diffusers)
pipe.to('cuda')


def generate(prompt):
    return pipe(prompt).images

gr.Interface(
    fn=generate,
    inputs=gr.Text(),
    outputs=gr.Gallery(),
).launch()

But AFAIK the diffusers pipeline stuff was still "processing"
https://github.com/huggingface/diffusers/pull/9982

hope it helps