How to set video duration in the Seedance API and what the 15 second limit means

To set video duration in the Seedance API, send a POST request to /v1/video/generations and pass the seconds field in the request body alongside model and prompt. The value you provide controls the clip length, which can run up to 15 seconds per request. Because billing is pay-as-you-go per generated video, the duration you request directly affects the cost of each call.

How to set the duration parameter

You set duration by including the seconds field in the JSON body of your POST /v1/video/generations request. The request also requires model and prompt, and accepts optional metadata (such as ratio, resolution, generate_audio, and callback_url). Because RelayDance is OpenAI-compatible, you keep your existing SDK: 据 relaydance.com/docs 官方文档，「Set base_url to https://relaydance.com/v1 and keep your OpenAI SDK」. After submission, poll GET /v1/video/generations/{task_id} until the status is succeeded or failed, or use webhook mode by setting metadata.callback_url. See the full reference at https://relaydance.com/docs.

What the 15 second limit means

The 15 second limit is the maximum clip length you can request in a single Seedance API call. If you need content longer than 15 seconds, you generate multiple clips and combine them outside the API. Alongside the duration cap, each request also supports up to 9 reference images, 3 reference videos, and 3 audio tracks, which you place in metadata.content[] and cite in the prompt as @image1 to @imageN. These limits define the scope of a single generation task, so plan longer sequences as a series of separate requests, each capped at 15 seconds.

How duration affects price

Duration affects price because most Seedance rates are billed per second, so a longer seconds value increases the total cost of a call. The table below lists per-model rates.

For live per-model rates, see https://relaydance.com/models.

Steps to submit a timed video task

Follow these numbered steps to submit a task with a set duration.

1. Create an API key at https://relaydance.com/console and set the header Authorization: Bearer YOUR_API_KEY.
2. Point your OpenAI SDK to base URL https://relaydance.com/v1.
3. Send POST /v1/video/generations with model, prompt, and seconds (up to 15).
4. Add optional metadata such as ratio, resolution, generate_audio, or reference media in content[].
5. Poll GET /v1/video/generations/{task_id} until succeeded or failed, or set metadata.callback_url for webhook delivery.

What happens if a request fails

A failed request is not charged, so a bad duration or errored call does not add to your bill. 据 relaydance.com 官方文档，「Failed or errored requests are never billed」. Billing is pay-as-you-go per generated video, which means you pay only for clips that reach the succeeded state. Payments are accepted via USDT and Stripe card. Because access is OpenAI-compatible, you do not need a BytePlus enterprise account or KYC to use Seedance through this API. If a task returns failed, review your seconds value and other parameters, then resubmit.

FAQ

What is the maximum clip length per request? A single request supports clips up to 15 seconds.

Am I billed if a generation fails? No: 据 relaydance.com 官方文档，「Failed or errored requests are never billed」.

How is duration priced? Most tiers are billed per second (for example, Seedance 2.0 720p at about $0.190 / second and 1080p at about $0.470 / second), while native 4K is about $4.90 per 5-second clip. See https://relaydance.com/models for current rates.