Dreamina-Seedance-2.5 - ByteDance AI Video Model (Global)
Dreamina-Seedance-2.5 is the global name for ByteDance Seedance 2.5. Build prompts and ModelArk API requests for 30-second 4K-ready, audio-synced video free.
Describe your scene — Seedance 2.5 generates up to 30 seconds of native audio-visual output.
Tip: put spoken dialogue in "double quotes" for better synced audio.
a cinematic scene. with natural ambient sound.
{
"model": "dreamina-seedance-2-5-260628",
"content": [
{
"type": "text",
"text": "a cinematic scene. with natural ambient sound."
}
],
"resolution": "720p",
"ratio": "16:9",
"duration": 10,
"generate_audio": true,
"camera_fixed": false,
"output_format": "mp4"
}Your files are processed locally in your browser. Nothing is uploaded to any server.
How to Use Dreamina-Seedance-2.5
Select the Model
Dreamina-Seedance-2.5 is the global name for Seedance 2.5. Use the model ID dreamina-seedance-2-5-260628 when calling the ModelArk API from outside China.
Compose the Shot
Write your prompt, pick ratio and resolution, set duration up to 30 seconds, and enable native audio or a fixed camera. The builder writes the request body.
Generate and Iterate
Copy the prompt or JSON and run it. Save the seed from a result you like so you can iterate on the same base with small, controlled changes.
Dreamina-Seedance-2.5 FAQ
What is Dreamina-Seedance-2.5?
Dreamina-Seedance-2.5 is the international product name for ByteDance's Seedance 2.5 video model, served through the ModelArk platform. The call name is dreamina-seedance-2-5-260628. In China the same model is offered as Doubao-Seedance-2.5.
What are its headline capabilities?
Dreamina-Seedance-2.5 generates a single continuous clip up to 30 seconds with synchronized native audio, supports text-to-video and image-to-video with first/last frame control, accepts up to 30 reference images plus reference video and audio, and outputs mp4 or high-color-precision mov.
Which resolutions does it support?
At launch it supports 480p and 720p, with 720p as the default and 1080p and 4K planned. You choose the resolution in the request, and the model generates natively at that size rather than upscaling afterward.