Qwen-image-edit-2511 mutilGPU error

[zhangquanwei962]

[rank6]: Traceback (most recent call last):
[rank6]:   File "/root/VideoX-Fun/examples/qwenimage/predict_t2i_edit_plus.py", line 234, in <module>
[rank6]:     sample = pipeline(
[rank6]:              ^^^^^^^^^
[rank6]:   File "/opt/conda/lib/python3.11/site-packages/torch/utils/_contextlib.py", line 116, in decorate_context
[rank6]:     return func(*args, **kwargs)
[rank6]:            ^^^^^^^^^^^^^^^^^^^^^
[rank6]:   File "/root/VideoX-Fun/videox_fun/pipeline/pipeline_qwenimage_edit_plus.py", line 864, in __call__
[rank6]:     noise_pred = self.transformer(
[rank6]:                  ^^^^^^^^^^^^^^^^^
[rank6]:   File "/opt/conda/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1739, in _wrapped_call_impl
[rank6]:     return self._call_impl(*args, **kwargs)
[rank6]:            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[rank6]:   File "/opt/conda/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1750, in _call_impl
[rank6]:     return forward_call(*args, **kwargs)
[rank6]:            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[rank6]:   File "/root/VideoX-Fun/videox_fun/utils/cfg_optimization.py", line 68, in wrapper
[rank6]:     cond_out = func(self, *args_i, **kwargs_i)
[rank6]:                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[rank6]:   File "/root/VideoX-Fun/videox_fun/models/qwenimage_transformer2d.py", line 1068, in forward
[rank6]:     encoder_hidden_states, hidden_states = block(
[rank6]:                                            ^^^^^^
[rank6]:   File "/opt/conda/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1739, in _wrapped_call_impl
[rank6]:     return self._call_impl(*args, **kwargs)
[rank6]:            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[rank6]:   File "/opt/conda/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1750, in _call_impl
[rank6]:     return forward_call(*args, **kwargs)
[rank6]:            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[rank6]:   File "/root/VideoX-Fun/videox_fun/models/qwenimage_transformer2d.py", line 598, in forward
[rank6]:     img_modulated, img_gate1 = self._modulate(img_normed, img_mod1, modulate_index)
[rank6]:                                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[rank6]:   File "/root/VideoX-Fun/videox_fun/models/qwenimage_transformer2d.py", line 573, in _modulate
[rank6]:     return x * (1 + scale_result) + shift_result, gate_result
[rank6]:            ~~^~~~~~~~~~~~~~~~~~~~
[rank6]: RuntimeError: The size of tensor a (1408) must match the size of tensor b (11264) at non-singleton dimension 1

RT. The inference will be error with 8 gpu. Can you give me some advice?