WAN and Hunyuan Video
(Example of WAN 2.1 i2v)
RuinedFooocus support generating videos with WAN 2.1 t2v, i2v and Hunyuan Video t2v.
The pipelines for these aren't as polished as for image generation but should hopefully work.
Image Number is used for the length of the video. For i2v the image put in Input image under PowerUp will be used.
You also need to create Performance settings: (These are examples, modify them to match your taste/hardware.)
- custom_steps: 20
- cfg: 5
- sampler_name: euler
- scheduler: simple
- clip_skip: 1
- custom_steps: 20
- cfg: 6.9
- sampler_name: uni_pc
- scheduler: simple
- clip_skip: 1
Using the default image size is probably not going to work. Here are some links to the recommended sizes: