Spaces:

ali-vilab

/

modelscope-text-to-video-synthesis

like 1.4k

Running on A10G

A lot of the time when I give it a prompt instead of trying to make a video of it, it just creates a blank wall of what looks like it's trying to be the text from the prompt and the camera may move a tiny bit but that's about it

Example:
I typed "war crimes" to see what would happen and this is what it gave me https://idontknowwhattonamemyself-text-to-video.hf.space/file=/tmp/a42f6713ae3cacb624e9b2caa2af37a681bb83fe/tmpuxtgphf5.mp4