Seedance 2.5 Launches: Native 30-Second Videos & 50 Multimodal References

nikola_mr64990 · x · 2026-07-31

ByteDance's AI video generation model, Seedance 2.5 (Dreamina), is now live, offering a powerful new tool for enterprise cross-lingual content creation.

The model natively supports up to 30 seconds of video generation, precise video editing, and up to 50 multimodal references to ensure consistency across visual concepts and multiple languages. Enterprise API access is available via BytePlus.

Related event: ByteDance Releases Seedance 2.5: Native 30s Video and Multimodal Control(60 posts)→

Original post →

More from Multimodal

Multimodal channel →