You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When trying to run any of the workflows for CharacterGen, get this error when trying to download the model using the force_download = true:
It begins to download, and files are in fact being downloaded. However, then:
!!! Exception during processing !!! [Errno 2] No such file or directory: 'E:\\SD\\NVIDIA\\ComfyUI\\Trellis-Hunyan\\Comfy3D_WinPortable\\ComfyUI\\custom_nodes\\ComfyUI-3D-Pack\\Checkpoints\\CharacterGen\\.cache\\huggingface\\download\\2D_Stage\\models\\checkpoint\\pytorch_model.bin.78c23c5e22d9c288d71414ec9604430b1298e652bb596ca2ff594792c1be4286.incomplete'
And download fails.
I have tried cloning the model repo locally and switching force_download to false. That does not work either, and produces this error:
[Comfy3D] Load CharacterGen MVDiffusion Model
_get_model_file() got an unexpected keyword argument 'resume_download'
What's the next step to fix this?
The text was updated successfully, but these errors were encountered:
This is because resume_download was depreciated, and finally removed, from huggingface_hub. It now auto resumes without prompting. resume_download needs to be removed from the each of the CharacterGen python scripts for it to work again.
A reported workaround is to downgrade the diffusers to 0.29.1, however I have not tested this myself.
@GabrielR thank you for raising this. I have posted a link to this thread from a related issue I posted on the Comfy 3D Pack repo issues. cc: @MrForExample
using Comfy30_WinPortable, Win 10, Nvidia 4080
When trying to run any of the workflows for CharacterGen, get this error when trying to download the model using the force_download = true:
It begins to download, and files are in fact being downloaded. However, then:
And download fails.
I have tried cloning the model repo locally and switching force_download to false. That does not work either, and produces this error:
What's the next step to fix this?
The text was updated successfully, but these errors were encountered: