Build a minimal Seedance 2.5 request and poll its result

Seedance 2.5 is available on RouterBase in text-to-video and image-to-video modes. A minimal integration needs an explicit payload and an asynchronous result handler. Here is a text-to-video request shape to begin with.

Explore RouterBase: https://routerbase.com/

Submit a single shot

Use bytedance/seedance-2-5-t2v. The following body specifies the ratio, resolution, duration, and audio choice rather than leaving those decisions implicit:

{
  "model": "bytedance/seedance-2-5-t2v",
  "prompt": "A ceramic cup on a wooden table. Steam rises slowly. The camera gently moves closer.",
  "ratio": "16:9",
  "resolution": "720p",
  "duration": 5,
  "generate_audio": false
}

Send it to POST https://routerbase.com/v1/videos/generations with authorization supplied from your server environment. The example describes a proposed request; no generated clip accompanies it.

Keep the job ID

The initial response returns an ID and a pending status. Save the ID, then request GET /v1/videos/generations/{id}. Check for a terminal status before accessing the video URL. On success, the documented response contains data[0].url; on failure, keep the error information available for diagnosis.

Give the interface a pending state and distinguish an unsuccessful submission from a failed generation. A timeout in your own client does not establish that the remote job failed. Retaining the ID lets you investigate the original job rather than blindly starting another one.

Change the mode deliberately

For image-to-video, change the model ID, provide one public HTTPS first-frame URL in image_urls, and use ratio: "adaptive". Choose 720p or 1080p; the 480p tier belongs to text-to-video. Crop the first frame before submitting if you need a particular shape. Do not send seed: RouterBase’s documentation says the upstream does not support it.

Match the request to the mode

RouterBase lists bytedance/seedance-2-5-t2v and bytedance/seedance-2-5-i2v. Both use /v1/videos/generations and return a pending task ID. Poll that endpoint with the ID until success or failure. Send ratio explicitly: text mode accepts fixed ratios, while image mode requires adaptive and exactly one public HTTPS image in image_urls. Both support 4–30 seconds and optional native audio.

Read the RouterBase implementation guide.

Disclosure: This article was drafted with AI assistance and reviewed against RouterBase's public documentation. The examples describe proposed requests; no generation test results are presented.