Spaces:

---

multimodalart

/

Chroma1-HD

like 18

Runtime error

I just duplicated this on my own zerogpu space with no changes and get the following error:

Runtime error
Exit code: 1. Reason: _model-0000(…): 87%|████████▋ | 6.81G/7.85G [00:06<00:00, 1.13GB/s]
transformer/diffusion_pytorch_model-0000(…): 100%|██████████| 7.85G/7.85G [00:07<00:00, 1.07GB/s]

(…)ion_pytorch_model.safetensors.index.json: 0%| | 0.00/107k [00:00<?, ?B/s]
(…)ion_pytorch_model.safetensors.index.json: 100%|██████████| 107k/107k [00:00<00:00, 4.53MB/s]

config.json: 0%| | 0.00/819 [00:00<?, ?B/s]
config.json: 100%|██████████| 819/819 [00:00<00:00, 7.99MB/s]

vae/diffusion_pytorch_model.safetensors: 0%| | 0.00/168M [00:00<?, ?B/s]
vae/diffusion_pytorch_model.safetensors: 100%|██████████| 168M/168M [00:00<00:00, 182MB/s]

Loading pipeline components...: 0%| | 0/5 [00:00<?, ?it/s]torch_dtype is deprecated! Use dtype instead!

Loading pipeline components...: 20%|██ | 1/5 [00:00<00:00, 228.29it/s]
Traceback (most recent call last):
File "/home/user/app/app.py", line 16, in
pipe = ChromaPipeline.from_pretrained(model_repo_id, torch_dtype=torch_dtype)
File "/usr/local/lib/python3.10/site-packages/huggingface_hub/utils/_validators.py", line 114, in _inner_fn
return fn(*args, **kwargs)
File "/usr/local/lib/python3.10/site-packages/diffusers/pipelines/pipeline_utils.py", line 1025, in from_pretrained
loaded_sub_model = load_sub_model(
File "/usr/local/lib/python3.10/site-packages/diffusers/pipelines/pipeline_loading_utils.py", line 849, in load_sub_model
loaded_sub_model = load_method(os.path.join(cached_folder, name), **loading_kwargs)
File "/usr/local/lib/python3.10/site-packages/transformers/modeling_utils.py", line 277, in _wrapper
return func(*args, **kwargs)
File "/usr/local/lib/python3.10/site-packages/transformers/modeling_utils.py", line 4974, in from_pretrained
model = cls(config, *model_args, **model_kwargs)
TypeError: T5EncoderModel.init() got an unexpected keyword argument 'offload_state_dict'

I initially wanted to modify it for lora support but somehow just a simple copy errors out? Any idea as for why?