我想直接调KOLORS模型，一直失败-。-请问应该用啥加载模型和CLIP啊，我试了几种一直报错

[doudoumei]

[doudoumei]

msdiffusion+story+ 2rolesin1img ,support txt2img and img2img cn只能用正方形图片吗，我改下了宽就报错了：
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "H:\ComfyUI_124_251\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1747, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "H:\ComfyUI_124_251\python_embeded\Lib\site-packages\diffusers\models\attention_processor.py", line 588, in forward
return self.processor(
^^^^^^^^^^^^^^^
File "H:\ComfyUI_124_251\ComfyUI\custom_nodes\ComfyUI_StoryDiffusion\msdiffusion\models\attention_processor.py", line 278, in call
custom_attention_masks = self.prepare_attention_mask_qk(boxes, phrase_idxes, hidden_states.shape[1],
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "H:\ComfyUI_124_251\ComfyUI\custom_nodes\ComfyUI_StoryDiffusion\msdiffusion\models\attention_processor.py", line 156, in prepare_attention_mask_qk
dummy_attention_mask = torch.clamp(dummy_attention_mask - box_mask, min=0)
~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~
RuntimeError: The size of tensor a (3840) must match the size of tensor b (3721) at non-singleton dimension 1

Prompt executed in 9.99 seconds

[smthemex]

[doudoumei]

我的意思是不用他的FACE参考功能，文件我都放好了，我想用STORY或者CLASSIC模式直接调可图模型-。-

[smthemex]

模式不兼容，没法做到。