Add diffusers weights
#4
by patrickvonplaten - opened
Make sure to set guidance scale to 1.0
from diffusers import AutoPipelineForText2Image
import torch
pipe = AutoPipelineForText2Image.from_pretrained("stabilityai/sdxl-turbo", torch_dtype=torch.float16, variant="fp16")
pipe.to("cuda")
prompt = "A cinematic shot of a baby racoon wearing an intricate italian priest robe."
steps = 1
image = pipe(prompt=prompt, num_inference_steps=steps, guidance_scale=0.0).images[0]
patrickvonplaten changed pull request status to open
does it work with img2img and controlnet pipelines?
Here a colab to try it out:
https://colab.research.google.com/drive/1yRC3Z2bWQOeM4z0FeJ0rF6fnDTTSdnAJ?usp=sharing
Yes, Image-to-Image works (see notebook)
@patrickvonplaten Thanks, I just tested it with CPU, and it takes 10 seconds to generate an image on Core i7. I will integrate it with FastSD CPU.
patrickvonplaten changed pull request status to merged