How to pick a model id on RelayDance: the full Seedance and happyhorse lineup

To pick a model id on RelayDance, open relaydance.com/models, choose the Seedance or HappyHorse model that matches your resolution and budget, and pass its exact id string in the model field of POST /v1/video/generations. For example, Seedance 2.5 at 1080p uses doubao-seedance-2-5-1080p at about $0.711 per second. Seedance Fast costs about $0.152 per second.

Which models are in the RelayDance lineup?

RelayDance offers ByteDance Seedance 2.0 and Seedance 1.5 Pro, a Seedance Fast tier, Seedance 2.5 at 1080p, a native 4K Seedance option, HappyHorse video models, and Grok Imagine video and image models. Grok text models and the gpt-image-2 image model are also listed. You select a model by passing its id string in the model field of your request. The published id for Seedance 2.5 at 1080p is doubao-seedance-2-5-1080p, added on 2026-10-10. For HappyHorse, Seedance 1.5 Pro, and the other tiers, copy the exact id from relaydance.com/models instead of guessing it, because that page lists the current ids and live per-model rates. In practice, picking an id comes down to three choices: model family, output resolution, and cost per second of video.

How do Seedance prices compare across models?

Seedance pricing rises with resolution and model version, from about $0.152 per second for Seedance Fast to about $4.90 per 5-second clip for native 4K.

A common pattern is to draft with Seedance Fast or Seedance 2.0 720p, then render final output at 1080p or native 4K. Rates can change, so confirm them on the models page before large batches.

How do I use the model id in an API request?

You place the model id in the model field of a standard OpenAI-compatible request to the RelayDance base URL.

1. Create an API key at relaydance.com/console and send it as Authorization: Bearer YOUR_API_KEY.
2. Point your client at RelayDance. According to the relaydance.com/docs, “Set base_url to https://relaydance.com/v1 and keep your OpenAI SDK”.
3. Copy the exact model id from the models page, for example doubao-seedance-2-5-1080p.
4. Send POST /v1/video/generations with model, prompt, seconds (clips up to 15 seconds), and metadata (ratio, resolution, generate_audio, callback_url, content[]).
5. Poll GET /v1/video/generations/{task_id} until status is succeeded or failed, or set metadata.callback_url to receive the final state by POST. The result contains the video url.

For images, use POST /v1/images/generations with an image model id such as gpt-image-2.

FAQ

The questions below cover billing, reference media, and account requirements.

Am I charged if a generation fails?
No. According to relaydance.com, “Failed or errored requests are never billed.” Billing is pay-as-you-go per generated video, and you can pay with USDT or a Stripe card.

How many reference files can one request include?
Each request accepts up to 9 reference images, 3 reference videos, and 3 audio tracks. Place them in metadata.content[] and cite them in the prompt as @image1 to @imageN.

Do I need a BytePlus enterprise account to use Seedance?
No. RelayDance provides direct OpenAI-compatible access to Seedance without a BytePlus enterprise account or KYC. You need a RelayDance API key and the base URL https://relaydance.com/v1.