Seedance 2.5

Seedance 2.5 is an AI video generator built for creators and production teams who want faster iteration from idea to finished clip. It’s a unified multimodal model that accepts text, image, video, and audio inputs and produces native video with synchronized audio in a single generation pass.

With Seedance 2.5 you can create native 4K videos (up to 3840 × 2160) up to 30 seconds in one run, reducing the need to stitch multiple short segments. You can also attach up to 50 multimodal reference assets—mixing images (identity/style), video (motion/composition cues), and audio (music or voice guidance)—to help orchestrate characters, products, or style into the same scene.

Seedance 2.5 is currently in preview, so pricing is not yet officially published. If you want early access, join the waitlist and try generating videos from a single prompt, with duration, resolution, and aspect ratio controls for different channels.

What is Seedance 2.5 used for?

Seedance 2.5 generates native AI videos for advertising, e-commerce, short-form storytelling, and synthetic data for embodied AI and autonomous driving.

What’s the maximum length of a single Seedance 2.5 generation?

You can generate up to 30 seconds of native video in a single run.

How many multimodal references can I attach with Seedance 2.5?

You can attach up to 50 reference assets per generation, in any mix of images, videos, and audio.

Does Seedance 2.5 generate synchronized audio?

Yes. Output includes a native audio track synchronized with the visuals.

What input and output formats does Seedance 2.5 use?

Inputs support text (UTF-8, Chinese & English), images (JPEG/PNG/WebP), video (MP4/MOV), and audio (WAV/MP3). Output is MP4 (H.264/H.265) with AAC synchronized audio, up to 30 seconds and up to 3840 × 2160.

Is Seedance 2.5 pricing available?

Seedance 2.5 is currently in preview and pricing hasn’t been officially published yet, so you’ll want to join the waitlist for updates when rates go live.