The number of frames to generate

The frame to start generating the video from

The number of inference steps to use for video generation