search

gokayfem / ComfyUI_VLM_nodes

Custom ComfyUI nodes for Vision Language Models, Large Language Models, Image to Music, Text to Music, Consistent and Random Creative Prompt Generation
Apache License 2.0
382 stars 31 forks source link

modelpatcher error #47

Closed nicolai256 closed 6 months ago

nicolai256 commented 6 months ago

Cannot import W:\visionsofchaos\Text To Image\ComfyUI\ComfyUI\custom_nodes\ComfyUI_VLM_nodes module for custom nodes: cannot import name 'ToImage' from 'torchvision.transforms.v2' (W:\visionsofchaos\Text To Image\ComfyUI\ComfyUI.venv\lib\site-packages\torchvision\transforms\v2__init__.py) Traceback (most recent call last): File "W:\visionsofchaos\Text To Image\ComfyUI\ComfyUI\nodes.py", line 1800, in load_custom_node module_spec.loader.exec_module(module) File "", line 883, in exec_module File "", line 241, in _call_with_frames_removed File "W:\visionsofchaos\Text To Image\ComfyUI\ComfyUI\custom_nodes\efficiency-nodes-comfyui__init__.py", line 1, in from .efficiency_nodes import NODE_CLASS_MAPPINGS File "W:\visionsofchaos\Text To Image\ComfyUI\ComfyUI\custom_nodes\efficiency-nodes-comfyui\efficiency_nodes.py", line 4, in from comfy.sd import ModelPatcher, CLIP, VAE ImportError: cannot import name 'ModelPatcher' from 'comfy.sd' (W:\visionsofchaos\Text To Image\ComfyUI\ComfyUI\comfy\sd.py)

gokayfem commented 6 months ago

update transformers library if you want to use this i think

click install pip packages inside ComfyUI Manager

write transformers>=4.38.2

then restart ComfyUI

image