Instructions to use Kijai/WanVideo_comfy with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusion Single File
How to use Kijai/WanVideo_comfy with Diffusion Single File:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
Error using FLASHVSR Wan Video Wrapper The size of tensor a (41472) must match the size of tensor b (36864) at non-singleton dimension 1
#101
by vhstudios - opened
While running the workflow I'm getting
The size of tensor a (41472) must match the size of tensor b (36864) at non-singleton dimension 1
How can i fix this?
Probably mismatch with the input video size/length and the generation resolution/frame count.
I ran into this issue too, but I figured it out!
Just don’t use the “Load Video (Upload)” node from ComfyUI-WanVideoWrapper—switch to the built-in “Load Video” node that comes with ComfyUI instead. That fixed it for me.
I asked an AI about it, and it’s probably because FlashVSR uses a lot of BF16 internally, which can cause compatibility issues. Hope this helps!