How to generate video with audio through an API

To generate video with audio through the RelayDance API, send a POST request to /v1/video/generations with a model, prompt, and seconds, then set metadata.generate_audio to enable audio output. Poll GET /v1/video/generations/{task_id} until the status is succeeded, and read the video URL from the result. The API is OpenAI-compatible, so you set the base URL to https://relaydance.com/v1 and reuse your existing SDK.

How do I submit a video generation request?

You submit a video task by sending POST /v1/video/generations with the model, prompt, and seconds fields, plus a metadata object for audio and other options. Set metadata.generate_audio to request audio, and use ratio, resolution, and callback_url as needed. According to the official relaydance.com/docs documentation, "Set base_url to https://relaydance.com/v1 and keep your OpenAI SDK" 「Set base_url to https://relaydance.com/v1 and keep your OpenAI SDK」据 relaydance.com 官方文档. Authenticate each request with an Authorization: Bearer YOUR_API_KEY header, using a key created at the RelayDance console. Clips can run up to 15 seconds per request.

How do I retrieve the finished video?

You retrieve the finished video by polling GET /v1/video/generations/{task_id} until the status is succeeded or failed, or by using webhook mode. When the status is succeeded, the response contains the video URL. For webhook delivery, set metadata.callback_url, and the final state is POSTed to your server. Failed tasks do not incur charges: according to the official relaydance.com documentation, "Failed or errored requests are never billed" 「Failed or errored requests are never billed」据 relaydance.com 官方文档. This billing rule applies across the pay-as-you-go model, so retrying an errored request carries no cost for the failed attempt.

How do I add reference media?

You add reference media by placing files in metadata.content[] and citing them in the prompt as @image1 to @imageN. A single request supports up to 9 reference images, 3 reference videos, and 3 audio tracks. These references let you condition generation on existing visuals or sound while the model produces the new clip. Follow these numbered steps:

1. Upload or reference media in metadata.content[].
2. Cite each item in the prompt as @image1 through @imageN.
3. Set metadata.generate_audio to include an audio track.
4. POST the request to /v1/video/generations.
5. Poll for status or wait for the callback.

What do the video models cost?

Pricing is pay-as-you-go per generated video, billed per second or per clip depending on the model. The table below lists current example rates; live per-model rates are published at relaydance.com/models.

Payments are accepted via USDT and Stripe card, and access requires no BytePlus enterprise account or KYC.

FAQ

Do I need a new SDK to use RelayDance? No. The API is OpenAI-compatible, so you set the base URL to https://relaydance.com/v1 and keep your existing OpenAI SDK.

Am I charged for failed requests? No. Failed or errored requests are never billed under the pay-as-you-go model.

How many reference files can one request include? A single request supports up to 9 reference images, 3 reference videos, and 3 audio tracks, with clips up to 15 seconds.