Model
1 entry · 11 June 2025
ByteDance said the text- and image-to-video model topped Artificial Analysis's leaderboards and generated a five-second 1080p clip in about 41 seconds.
Models & capabilities