stabilityai/stable-diffusion-3-medium-diffusers · pipeline参数不足报错，我用的是示例代码

Jun 27

Pipeline <class 'diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline'> expected {'vae', 'unet', 'image_encoder', 'text_encoder', 'feature_extractor', 'tokenizer', 'safety_checker', 'scheduler'}, but only {'vae', 'scheduler', 'tokenizer', 'text_encoder'} were passed.

GoodResearch

Jul 8

Did you solve it? I am also getting this error.

kyxkkk

Jul 10

Did you solve it? I am also getting this error

Jiaozi123

Jul 15

Me too

maxwell60701

Jul 19

•

edited Jul 19

maybe you should generate a token with permission "write",then pass the token parameter,（示例代码应该跑不通的）
by the way,set your own cache_dir if you do not have enough space on disk C,the model is nearly 14G,

import torch
from diffusers import StableDiffusion3Pipeline


pipe = StableDiffusion3Pipeline.from_pretrained("stabilityai/stable-diffusion-3-medium-diffusers", torch_dtype=torch.float16,cache_dir="<your_directory>",token=<your_token>)
pipe = pipe.to("cuda")

image = pipe(
    "A cat holding a sign that says hello world",
    negative_prompt="",
    num_inference_steps=28,
    guidance_scale=7.0,
).images[0]
image

jingyiwu

Jul 24

Did you solve it? I am also getting this error.
Have you solved it?I am also getting this error.