ByteDance's Seedance 2.5 Lands on Dreamina with 30s Videos and 50 Multimodal References

HeyNayeem · x · 2026-08-02

ByteDance's AI creation platform Dreamina has officially launched the Seedance 2.5 video generation model. The major upgrades include the ability to load up to 50 multimodal reference assets (images, videos, and audio) into a single generation, and support for up to 30-second continuous video generation for highly coherent storytelling. Paired with its native editing workflow and creator-friendly pricing, the platform offers a highly cost-effective route for production-ready AI video creation.

Related event: ByteDance Seedance 2.5: 50 refs, 30s video, CapCut integration(33 posts)→

Original post →

More from Multimodal

Multimodal channel →