Wan 2.1 I2V & T2V Workflows - 12 GB - Low VRAM - GGUF

These are the workflows I've been using for Wan so far.

They are heavily based upon this workflow by oscarchuncha654 so all credit to them:

https://civitai.com/articles/12202/wan-21-480-gguf-q5-model-on-low-vram-8gb-and-16-gb-ram-fastest-workflow-10-minutes-max-now-8-mins

You'll need CUDA, Triton and SageAttention.

I used the below guide for CUDA:

https://medium.com/@fernandopalominocobo/installing-cuda-for-pytorch-easily-explained-windows-users-4d3b7db5f2e0

And this for Triton and SageAttention:

https://www.reddit.com/r/StableDiffusion/comments/1j0enkx/automatic_installation_of_triton_and/

For the sampler I tend to switch between euler and uni_pc.