Phi3Transformer does not support an attention implementation

[mixoadrian]

I do not know how to make sense of the error so i paste it here:

Fetching 10 files: 100%|███████████████████████████████████████████████████████████████| 10/10 [04:55<00:00, 29.52s/it]
!!! Exception during processing !!! Phi3Transformer does not support an attention implementation through torch.nn.functional.scaled_dot_product_attention yet. Please request the support for this architecture: huggingface/transformers#28005. If you believe this error is a bug, please open an issue in Transformers GitHub repository and load your model with the argument attn_implementation="eager" meanwhile. Example: model = AutoModel.from_pretrained("openai/whisper-tiny", attn_implementation="eager")
Traceback (most recent call last):
File "E:\comfy\ComfyUI_windows_portable\ComfyUI\execution.py", line 323, in execute
output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\comfy\ComfyUI_windows_portable\ComfyUI\execution.py", line 198, in get_output_data
return_values = _map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\comfy\ComfyUI_windows_portable\ComfyUI\execution.py", line 169, in map_node_over_list
process_inputs(input_dict, i)
File "E:\comfy\ComfyUI_windows_portable\ComfyUI\execution.py", line 158, in process_inputs
results.append(getattr(obj, func)(**inputs))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\comfy\ComfyUI_windows_portable\ComfyUI\custom_nodes\OmniGen-ComfyUI_init.py", line 116, in gen
pipe = OmniGenPipeline.from_pretrained(omnigen_dir)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\comfy\ComfyUI_windows_portable\ComfyUI\custom_nodes\OmniGen-ComfyUI\OmniGen\pipeline.py", line 82, in from_pretrained
model = OmniGen.from_pretrained(model_name)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\comfy\ComfyUI_windows_portable\ComfyUI\custom_nodes\OmniGen-ComfyUI\OmniGen\model.py", line 197, in from_pretrained
model = cls(config)
^^^^^^^^^^^
File "E:\comfy\ComfyUI_windows_portable\ComfyUI\custom_nodes\OmniGen-ComfyUI\OmniGen\model.py", line 186, in init
self.llm = Phi3Transformer(config=transformer_config)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\comfy\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\models\phi3\modeling_phi3.py", line 948, in init
super().init(config)
File "E:\comfy\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 1388, in init
config = self._autoset_attn_implementation(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\comfy\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 1565, in _autoset_attn_implementation
config = cls._check_and_enable_sdpa(
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\comfy\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 1731, in _check_and_enable_sdpa
raise ValueError(
ValueError: Phi3Transformer does not support an attention implementation through torch.nn.functional.scaled_dot_product_attention yet. Please request the support for this architecture: huggingface/transformers#28005. If you believe this error is a bug, please open an issue in Transformers GitHub repository and load your model with the argument attn_implementation="eager" meanwhile. Example: model = AutoModel.from_pretrained("openai/whisper-tiny", attn_implementation="eager")

Prompt executed in 298.22 seconds

[thephimart]

Go to \ComfyUI_windows_portable\ComfyUI\models\AIFSH\Shitao\OmniGen-v1
Open config.json
Scroll to the very bottom
change this:
"_attn_implementation": "sdpa"
to this:
"_attn_implementation": "eager"

worked for me hope it helps

[emimix]

Go to \ComfyUI_windows_portable\ComfyUI\models\AIFSH\Shitao\OmniGen-v1 Open config.json Scroll to the very bottom change this: "_attn_implementation": "sdpa" to this: "_attn_implementation": "eager"

worked for me hope it helps

That fixed it for me...thank you!

[xgs87762GH]

Go to \ComfyUI_windows_portable\ComfyUI\models\AIFSH\Shitao\OmniGen-v1 Open config.json Scroll to the very bottom change this: "_attn_implementation": "sdpa" to this: "_attn_implementation": "eager"

worked for me hope it helps

解决了，感谢！