How to call the Seedance 2.0 video generation API

To call the Seedance 2.0 video generation API, submit a POST request to https://relaydance.com/v1/video/generations with your API key in the Authorization header, including required parameters: model (set to the Seedance 2.0 identifier), prompt, seconds (video duration), and optional metadata like ratio, resolution, and callback_url. RelayDance is an OpenAI-compatible API that provides access to ByteDance's Seedance 2.0 and other advanced video models through a unified interface.

Setting up authentication

Authentication uses bearer token format. Create an API key at https://relaydance.com/console, then include it in every request using the Authorization header: Authorization: Bearer YOUR_API_KEY. Replace YOUR_API_KEY with your actual key from the console. All requests must use HTTPS to the base URL https://relaydance.com/v1. Keep your API key private and never commit it to version control. You can create multiple keys for different applications or environments.

Submitting a video generation request

Send a POST request to /v1/video/generations with a JSON body containing your parameters. The model parameter should specify Seedance 2.0. The prompt describes what video to generate. The seconds parameter sets the video length. Optional metadata fields include ratio (aspect ratio), resolution, generate_audio (whether to create audio), callback_url (for webhook notifications), and content array (for reference media). You can reference up to 9 images, 3 videos, and 3 audio tracks per request, cited in your prompt as @image1, @image2, etc.

Monitoring generation status

After submission, poll the status endpoint GET /v1/video/generations/{task_id} until the response shows status as "succeeded" or "failed". The task_id is returned from your initial POST request. Alternatively, use webhook mode by setting metadata.callback_url, and RelayDance will POST the final result to your server automatically. When status is "succeeded", the response contains the video_url where your generated video is hosted. Check https://relaydance.com/models for current per-model pricing and availability.

Working with reference media

Reference media enables more precise control over generated videos. Include media files in the metadata.content array as objects with type and url. In your prompt, reference them by position using @image1, @image2, and so on. Each reference clip can be up to 15 seconds. This allows you to provide style references, objects, scenes, or visual elements that guide the generation. Audio tracks can be referenced to create videos with specific soundscapes or music. The API supports combining multiple reference types in a single request for complex creative outputs.

Pricing and billing

RelayDance offers competitive pricing through both USDT and Stripe card payments. Mainstream model groups, including Seedance models, are priced 30 percent or more below official list prices. Failed or errored requests are never billed, so you only pay for successful generations. View live per-model rates at https://relaydance.com/models to see exact pricing for Seedance 2.0 and other available models. Rates may vary based on parameters like resolution and duration.

Frequently asked questions

1. What models are available besides Seedance 2.0? RelayDance provides access to ByteDance Seedance 1.5 Pro, HappyHorse video models, Grok Imagine video and image models, and Grok text models, all through the same OpenAI-compatible API interface.
2. How long does a video generation take? Generation time depends on video length and complexity. Poll the status endpoint or use webhooks to be notified when generation completes. Check https://relaydance.com/docs for detailed documentation on timing expectations.
3. Can I generate images with this API? Yes, image generation uses the endpoint POST /v1/images/generations with similar parameters and authentication, supporting Grok Imagine and other image models available on the platform.