Google video model returns to top: Gemini Omni Flash tops Video Arena

robot
Abstract generation in progress
According to Beat Insight monitoring, Google DeepMind’s newly released video generation model Gemini Omni Flash has topped the Video Arena blind testing leaderboard with an Elo score of 1404. The new model leads the previously dominant ByteDance Seedance 2.0 Mini by a full 101 points, setting a new record for the largest score gap on this leaderboard. As a preference ranking based on human blind test voting, Video Arena had previously been consistently occupied by ByteDance’s Seedance series models in the top three. Among them, Seedance 2.0 Mini took first place with 1303 points thanks to better multi-shot visual stability and faster generation speed. Gemini Omni Flash’s rise to the top this time marks that Google’s video generation models have, within half a year, overtaken competitors such as ByteDance—jumping from the lagging position in the Veo era to a leading one.
View Original
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
  • Reward
  • Comment
  • Repost
  • Share
Comment
Add a comment
Add a comment
No comments
  • Pinned