Seedance 1.5 Pro

Seedance 1.5 Pro is ByteDance Seed's audio-visual generation model, launched December 16, 2025. Unlike most video models that generate silent clips, it produces synchronized audio and video in one pass using a dual-branch Diffusion Transformer with a cross-modal joint module, trained with SFT and RLHF on multi-dimensional reward models.

The model supports 4 to 12 second clips with millisecond-precision audio sync, native multilingual dialogue including Chinese dialects, and cinema-grade camera moves such as Hitchcock zooms. It is available on Volcano Engine (model ID Doubao-Seedance-1.5-pro), Jiemeng AI, and Doubao, with free trials on CapCut Desktop. Seedance 2.x successors arrived later; this entry covers the 1.5 generation. No public weights exist.