Will the text-to-video model be released?

[jianzongwu]

I'm wondering when will the text-to-video model mentioned in the technical report be released in this repo. I know one can generate an image using T2I models and then use the I2V model to convert it to a video. But this process cannot guarantee the motion of the video is following the instructions people want. You can't specify "a dog walking from left to right" in a T2I image, right?

[18445864529]

I doubt whether they'll open-source it at all 👎