This is a video frame image provider + indexer/video creation nodes for hooking up to iterators and ranges and ControlNets and such for invokeAI node experimentation. Think animation + ControlNet outputs.
A high number of frames may cause a bit of time to create with the example workflow. I'd stick to smaller batches of frames at first and slowly ramp up to find a length of time you're willing to wait for the task to start/complete.
Render every other frame and then interpolate the in-between frames with something like RIFE or FILM for some more smoothing and a lot less rendering time.
Feel free to contribute or take ispiration for your own ideas and thanks to those who already have!