Video is already possible as is shown by e.g. SALT_VERSE [1] (which may or may not use SD/DALL-E/... but that is not that interesting), what is not there yet is a --txt2vid script option. Implementing this would not be that hard but the processing time needed is quite substantial.
Something else which would be possible is the use of a model like SD in combination with a frame interpolation model like [2] as a video generator. Use SD to generate key frames, feed these to FILM and let it generate the intermediate frames and you should get video.
Something else which would be possible is the use of a model like SD in combination with a frame interpolation model like [2] as a video generator. Use SD to generate key frames, feed these to FILM and let it generate the intermediate frames and you should get video.
[1] https://twitter.com/SALT_VERSE
[2] https://film-net.github.io/