How is Seedance 2.5 different from Seedance 2.0?

On the numbers Reelfo reads from its model registry: up to thirty reference images a shot against nine, up to ten audio references, and clips up to thirty seconds against fifteen. Both make sound with the picture. Seedance 2.5 costs more per second, so use it where the extra references or length matter.

Does Seedance 2.5 make sound?

Yes — dialogue, ambience and effects come with the clip, and the clip is charged at the same price either way. If you do not want sound, turn Sound off in the reel settings.

What is in the sample on this page?

A real five-second clip made with Seedance 2.5 from a still, 720p, with sound. The still was drawn with an image model; the motion and the sound are Seedance 2.5’s.

new

Seedance 2.5

Seedance 2.5 by ByteDance is the larger sibling of Seedance 2.0: more reference images a shot, audio references that can steer a voice or a beat, and clips that run up to thirty seconds. It makes its sound with the picture, and it runs on Reelfo as a managed API call — nothing to install, no hardware to rent.

制作をはじめるショーケースを見る

イメージモデル

スペック

モデル仕様

最大解像度

720p

最大の長さ

30s

速度

~60s per clip

クレジット単価

2365 credits / 5s clip

できること

対応機能

AI Text to Video Generator

Reference to Video

仕組み

主な機能

Thirty References a Shot

Up to thirty reference images in one clip — characters, places, props — each tagged in the prompt, so a crowded scene can be held to your own cast instead of whoever the model imagines.

Sound Made With the Picture

Dialogue, ambience and effects are generated with the clip, and up to ten audio references can steer a voice or a beat. Turn sound off in the settings and the clip is asked for without it.