pytorch diffusers text-to-image

Chinese Latent Diffusion Model

我们开源了一个中文 Lattent Diffusion 模型(开放领域通用版本,768 分辨率)

from diffusers import StableDiffusionPipeline

model_id = "alibaba-pai/pai-diffusion-general-xlarge-zh"
pipe = StableDiffusionPipeline.from_pretrained(model_id)
pipe = pipe.to("cuda")

prompt = "雾蒙蒙的日出在湖面上"
image = pipe(prompt).images[0]  
image.save("result.png")