I encountered an error when executing CLIPVisionLoader in ComfyUI. The error suggests a size mismatch in the vision model embeddings.
Error message:
Error occurred when executing CLIPVisionLoader: Error(s) in loading state_dict for CLIPVisionModelProjection: size mismatch for vision_model.embeddings.position_embedding.weight: copying a param with shape torch.Size([577, 1024]) from checkpoint, the shape in current model is torch.Size([257, 1024]).
Full traceback:
File "[PATH_TO_COMFYUI]\ComfyUI\execution.py", line 151, in recursive_execute
output_data, output_ui = get_output_data(obj, input_data_all)
File "[PATH_TO_COMFYUI]\ComfyUI\execution.py", line 81, in get_output_data
return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
File "[PATH_TO_COMFYUI]\ComfyUI\execution.py", line 74, in map_node_over_list
results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
File "[PATH_TO_COMFYUI]\ComfyUI\nodes.py", line 892, in load_clip
clip_vision = comfy.clip_vision.load(clip_path)
File "[PATH_TO_COMFYUI]\ComfyUI\comfy\clip_vision.py", line 117, in load
return load_clipvision_from_sd(sd)
File "[PATH_TO_COMFYUI]\ComfyUI\comfy\clip_vision.py", line 101, in load_clipvision_from_sd
m, u = clip.load_sd(sd)
File "[PATH_TO_COMFYUI]\ComfyUI\comfy\clip_vision.py", line 46, in load_sd
return self.model.load_state_dict(sd, strict=False)
File "[PATH_TO_COMFYUI]\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 2189, in load_state_dict
raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format(
neburagi
commented
1 month ago
I encountered an error when executing CLIPVisionLoader in ComfyUI. The error suggests a size mismatch in the vision model embeddings.
Error message: Error occurred when executing CLIPVisionLoader: Error(s) in loading state_dict for CLIPVisionModelProjection: size mismatch for vision_model.embeddings.position_embedding.weight: copying a param with shape torch.Size([577, 1024]) from checkpoint, the shape in current model is torch.Size([257, 1024]).
Full traceback: File "[PATH_TO_COMFYUI]\ComfyUI\execution.py", line 151, in recursive_execute output_data, output_ui = get_output_data(obj, input_data_all) File "[PATH_TO_COMFYUI]\ComfyUI\execution.py", line 81, in get_output_data return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True) File "[PATH_TO_COMFYUI]\ComfyUI\execution.py", line 74, in map_node_over_list results.append(getattr(obj, func)(**slice_dict(input_data_all, i))) File "[PATH_TO_COMFYUI]\ComfyUI\nodes.py", line 892, in load_clip clip_vision = comfy.clip_vision.load(clip_path) File "[PATH_TO_COMFYUI]\ComfyUI\comfy\clip_vision.py", line 117, in load return load_clipvision_from_sd(sd) File "[PATH_TO_COMFYUI]\ComfyUI\comfy\clip_vision.py", line 101, in load_clipvision_from_sd m, u = clip.load_sd(sd) File "[PATH_TO_COMFYUI]\ComfyUI\comfy\clip_vision.py", line 46, in load_sd return self.model.load_state_dict(sd, strict=False) File "[PATH_TO_COMFYUI]\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 2189, in load_state_dict raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format(
Environment: