SVD error

[workproduct]

New install of Comfy UI + Comfy UI manager

ran the update .bat that includes python update

installed the SVD models

using the second example workflow from here

https://comfyanonymous.github.io/ComfyUI_examples/video/

adjusted the paths on the loaders

Getting the following error

`Error occurred when executing CLIPTextEncode:

'NoneType' object has no attribute 'tokenize'

File "E:_SD\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\execution.py", line 153, in recursive_execute output_data, output_ui = get_output_data(obj, input_data_all) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:_SD\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\execution.py", line 83, in get_output_data return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:_SD\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\execution.py", line 76, in map_node_over_list results.append(getattr(obj, func)(**slice_dict(input_data_all, i))) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:_SD\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\nodes.py", line 55, in encode tokens = clip.tokenize(text) ^^^^^^^^^^^^^`

[workproduct]

the first load check point was the problem, needed to not be set to the xt model. Went with 1.5 and now its fine.

[amir272]

@workproduct Can you please explain what is 1.5? I am getting the same error. I am using svd.safetensor as the checkpoints for two stages

[ltdrdata]

You have to use ImageOnlyCheckpointLoader and SVD_img2vid_conditioning for svd checkpoint. SVD checkpoint doesn't provides CLIP and VAE.

[amir272]

@ltdrdata Thank you, this looks fine as of now.