Today we officially launch Seedance 2.5, the new‑generation video creation model. Since Seedance 2.0, users demand more than a clip; they want a complete story. Building on a unified multimodal audio‑video joint‑generation architecture, Seedance 2.5 focuses on foundational and reference‑based generation, delivering breakthroughs in long‑form storytelling, multimodal reference, and editing.

A key highlight is 30 seconds per generation with multi‑round extensions, SMART shot transitions तेल, and higher image, audio, and motion quality. Users can now input up to 30 images, 10 video clips, and 10 audio clips in one pass, enabling complex ideas across scenes and subjects. The model also offers timestamp‑level control for precise editing, green‑screen, perspective shifts, and reference‑based adjustments.

Seedance 2.5 is rolling out on Jimeng AI, Doubao Pro, and other platforms, with API access coming soon via Byte Plus Model Ark. Try it, share feedback, and explore its cinematic‑quality outputs.