Sora
OpenAI, always with audio and lip-sync
Sora 2 by OpenAI — flagship with native audio, dialog and lip-sync. Base 720p ($0.10/sec) and Pro up to true 1080p ($0.30+/sec). Durations 4/8/12/16/20 seconds.
Model versions
Sora 2
Sora 2 by OpenAI: state-of-the-art video with native audio and lip-sync out of the box. 720p, durations 4/8/12/16/20 sec. $0.10/sec.
18 tok./sec
Sora 2 Pro
Pro variant of Sora 2: higher detail and smoother motion. Up to true 1080p. Durations 8/12/16/20 sec, with audio. $0.30/sec @ 720p, higher resolutions cost more.
54 tok./sec