ByteDance·closed·Proprietary
Dreamina Seedance 2.0
ByteDance's multimodal Seedance 2.0 text/image-to-video model (Dreamina / BytePlus ModelArk), with native audio and 4–15s clips up to 4K. Artificial Analysis lists ~$9.07/min (~$0.151/sec) for default 1080p API settings.
- Max res 3840 × 2160
- Max duration 15s
- Price $0.15 / s
- Arena Elo 1,225 · #3/17
- Speed —
Specs
- License
- Proprietary
- Open weights
- No
- Max resolution
- 3840 × 2160
- Max duration
- 15s
- Generation
- —
- Release
- Feb 12, 2026
Benchmarks
Artificial Analysis Video Arena Elo
Blind pairwise preference Elo for text-to-video models from the Artificial Analysis Video Arena (with-audio leaderboard). Scores are only comparable within the same Arena protocol and voter pool. (source)
1,225
#3 of 17