module diffusers has no attribute HiDreamImagePipeline

#6
by khanakia - opened

I am using this example

from diffusers import DiffusionPipeline

pipe = DiffusionPipeline.from_pretrained("HiDream-ai/HiDream-I1-Dev")

prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]

and getting error

(.venv) ➜  huggingface-course git:(main) ✗ python diffuser2.py
Traceback (most recent call last):
  File "/Volumes/D/www/ai/huggingface-course/diffuser2.py", line 3, in <module>
    pipe = DiffusionPipeline.from_pretrained("HiDream-ai/HiDream-I1-Dev")
  File "/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/huggingface_hub/utils/_validators.py", line 114, in _inner_fn
    return fn(*args, **kwargs)
  File "/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/diffusers/pipelines/pipeline_utils.py", line 773, in from_pretrained
    cached_folder = cls.download(
        pretrained_model_name_or_path,
    ...<14 lines>...
        **kwargs,
    )
  File "/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/huggingface_hub/utils/_validators.py", line 114, in _inner_fn
    return fn(*args, **kwargs)
  File "/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/diffusers/pipelines/pipeline_utils.py", line 1473, in download
    pipeline_class = _get_pipeline_class(
        cls,
    ...<7 lines>...
        revision=custom_revision,
    )
  File "/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/diffusers/pipelines/pipeline_loading_utils.py", line 427, in _get_pipeline_class
    pipeline_cls = getattr(diffusers_module, class_name)
  File "/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/diffusers/utils/import_utils.py", line 813, in __getattr__
    raise AttributeError(f"module {self.__name__} has no attribute {name}")
AttributeError: module diffusers has no attribute HiDreamImagePipeline

We recommend to install initialize pipeline following this command:

import torch
from transformers import PreTrainedTokenizerFast, LlamaForCausalLM
from diffusers import HiDreamImagePipeline
tokenizer_4 = PreTrainedTokenizerFast.from_pretrained("meta-llama/Meta-Llama-3.1-8B-Instruct")
text_encoder_4 = LlamaForCausalLM.from_pretrained(
    "meta-llama/Meta-Llama-3.1-8B-Instruct",
    output_hidden_states=True,
    output_attentions=True,
    torch_dtype=torch.bfloat16,
)

pipe = HiDreamImagePipeline.from_pretrained(
    "HiDream-ai/HiDream-I1-Full",  # "HiDream-ai/HiDream-I1-Dev" | "HiDream-ai/HiDream-I1-Fast"
    tokenizer_4=tokenizer_4,
    text_encoder_4=text_encoder_4,
    torch_dtype=torch.bfloat16,
)

pipe = pipe.to('cuda')

Getting this error if i run above code

ImportError: cannot import name 'HiDreamImagePipeline' from 'diffusers' (/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/diffusers/__init__.py)

maybe we are using the not compatible version of diffusers ? i got the same error :(

Which version does work?

I am using the latest version of diffuser https://github.com/huggingface/diffusers
version = "0.33.1"

Screenshot 2025-04-20 at 10.19.23 PM.png

Apologies for the late reply! Our model has not yet been merged into the latest stable release of Diffusers (v0.33.1). Please follow the Inference with Diffusers section in our github repository to install Diffusers from source code for better compatibility:

pip install git+https://github.com/huggingface/diffusers.git

When will the latest stable release of Diffusers contain the needed code for iDreamImageTransformer2DModel and HiDreamImagePipeline?

This would depend on the official Diffusers release schedule, which we don't directly control. We’ll keep an eye on updates from the Diffusers team and will aim to support the next stable version as soon as possible.

khanakia changed discussion status to closed
Your need to confirm your account before you can post a new comment.

Sign up or log in to comment