Hunyuan Video Generation Large Model Video-to-Video (V2V) ComfyUI HunyuanVideoWrapper
 
ComfyUI HunyuanVideoWrapper

The memory usage completely depends on the resolution and frame count; even on 24GB, don't expect to reach very high levels.

The good news is that the model can create functional videos even at very low resolutions.

Please note that the frame count reading upper limit should not be arbitrarily modified, otherwise an error will occur.
Ensure that the final input frame count to the sampler 1 is divisible by 4. For example, 101 1 = 100 can be divisible by 4.