ModelsBusiness

ByteDance’s Volcano Engine unveils Seedance 2.5 — 30‑second AI videos without stitching

Volcano Engine announces Seedance 2.5, a video model that generates contiguous clips up to 30 seconds long and will launch in early July.

In detail

  • Seedance 2.5 creates single clips up to 30 seconds without post‑stitching.
  • Can handle up to 50 additional inputs simultaneously (reference images, audio, etc.).
  • Generated videos are editable afterwards while preserving visual style and look.
  • Volcano Engine also announced Doubao 2.1 Pro (language model), Seedream 5.0 Pro (image), Seed‑Audio 1.0; Doubao 2.1 Pro is claimed to cost ~80% less than Claude Opus 4.6.

Why it matters

Longer coherent video generation and rich multi‑input support reduce effort for marketing and creative workflows; lower cost claims for enterprise LLMs may pressure vendor pricing in NLP services.

For you Assess Seedance 2.5 for marketing or product video pilots planned for July; request demos to validate fidelity and post‑edit workflow with your brand assets.

← All news

Summaries are generated automatically and link to the original source.