Which video model to use: Seedance 2.0 vs 1.5 Pro vs HappyHorse vs Grok Imagine

RelayDance offers Seedance 2.0, Seedance 1.5 Pro, HappyHorse video models, and Grok Imagine video and image models through one OpenAI-compatible API. Choose Seedance 2.0 for graded resolution tiers (720p at about $0.190/second, 1080p at about $0.470/second, native 4K at about $4.90 per 5-second clip), Seedance Fast at about $0.152/second for cheaper runs, and Grok Imagine for image tasks where output is free and only input is billed. All models share one base URL and pay-as-you-go billing.

What are the price differences across Seedance tiers?

Seedance pricing scales with resolution, so pick the tier that matches your delivery target. On relaydance.com/models, Seedance 2.0 720p costs about $0.190/second, 1080p costs about $0.470/second, and native 4K costs about $4.90 per 5-second clip. Seedance Fast is priced at about $0.152/second, the lowest per-second rate listed, which suits iteration and drafts. Because billing is pay-as-you-go per generated video, the price you pay follows the seconds and resolution you request. Failed requests carry no cost: 「Failed or errored requests are never billed」 (据 relaydance.com 官方文档). Live per-model rates are published at relaydance.com/models.

How do the models compare?

Use this table to match each model family to a use case and its listed rate.

Seedance 1.5 Pro and HappyHorse are additional video model families available through the same API; check relaydance.com/models for their current rates.

When should you use Grok Imagine for images?

Choose Grok Imagine (gpt-image-2) when you need image generation rather than video. According to relaydance.com/models, image output is free and only the input is billed, so 4K and 1K output cost the same. Image-to-image starts from about CNY 0.035. This billing shape means resolution does not raise the output charge, which changes how you plan batches compared with the per-second video models. Image tasks use POST /v1/images/generations, while video tasks use POST /v1/video/generations. Both endpoints share the same authentication and the same pay-as-you-go structure, so you can mix image and video calls under one API key.

How do you connect and submit a job?

Connect with your existing OpenAI SDK and point it at the RelayDance base URL. According to the relaydance.com/docs, 「Set base_url to https://relaydance.com/v1 and keep your OpenAI SDK」. Follow these steps:

1. Create an API key at https://relaydance.com/console and send it as Authorization: Bearer YOUR_API_KEY.
2. Set base_url to https://relaydance.com/v1.
3. Submit a video task with POST /v1/video/generations, including model, prompt, seconds, and metadata (ratio, resolution, generate_audio, callback_url, content[]).
4. Poll GET /v1/video/generations/{task_id} until status is succeeded or failed, or set metadata.callback_url for webhook delivery.

Reference media go in metadata.content[] and are cited as @image1 to @imageN. Details are documented at relaydance.com/docs.

FAQ

Are failed requests charged? No. 「Failed or errored requests are never billed」 (据 relaydance.com 官方文档), so only successful generations count against your pay-as-you-go balance.

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

Which payment methods are supported? USDT and Stripe card. Access is direct OpenAI-compatible access to Seedance with no BytePlus enterprise account or KYC required.