Seedance 2.0 holds #1 on Artificial Analysis Video Arena across text-to-video and image-to-video

ByteDance's Seedance 2.0 currently sits at #1 on the Artificial Analysis Video Arena leaderboard across both text-to-video (Elo 1,269) and image-to-video (Elo 1,351) — ahead of Kling 3.0, Veo 3.1, and the now-deprecated Sora 2.

The dual-leaderboard sweep is notable. Seedance 2.0's twelve-input architecture (nine images + three video clips + three audio files) is what differentiates it operationally — it's built for production workflows that combine references and stylization, not for raw photorealism comparisons.

The Elo leadership has held since Seedance 2.0's February release, which is a long stable run by video-model standards. Whether the lead survives the next Veo or Kling release is the open question; Kling 3.0 closed the gap in April and would have caught Seedance had Kling executed a hypothesized but unannounced 3.1 release on schedule.

Artificial Analysis — Video Arena → · Atlas Cloud — video model comparison →