Seedance 2.5 Launches with 30s Native Video Generation and Multimodal Inputs

HeyNayeem · x · 2026-08-05

ByteDance's AI creative platform Lumina AI has officially launched the Seedance 2.5 video generation model. The new version significantly upgrades its capabilities, supporting up to 30 seconds of native video generation in a single pass and accepting up to 50 multimodal reference inputs. It also offers finer video editing controls and native support for over 10 languages. Additionally, the platform introduced an exclusive '3D Director's Desk' feature, allowing users to generate white models from uploaded images. To celebrate the launch, Lumina is offering promotional discounts, including 68 free credits and a $0.99 starter pack for new users.

Original post →

More from Multimodal

Multimodal channel →