AI Video Generation API: Veo 3.1, Seedance, Grok Imagine, One Endpoint
TL;DR: GenRelay unifies the top text-to-video, image-to-video, and video-editing models — Google Veo 3.1, Seedance 2.0, Grok Imagine 1.5, and Gemini Omni Flash — behind one OpenAI-compatible API. Every Veo 3.1 generation is a fixed 8-second clip at $0.06, works out to $0.0075/second.
Supported video models
| Model | Provider | Price | Notes |
|---|---|---|---|
| Seedance 2.0 Mini | ByteDance | $0.04 / second | Fastest, lowest cost per second |
| Seedance 2.0 Fast | ByteDance | $0.06 / second | Higher quality tier |
| Veo 3.1 | $0.06 / generation | Fixed 8-second clip, text/image/reference-to-video | |
| Gemini Omni Flash | $0.10 / generation | Video with native audio | |
| Grok Imagine 1.5 | xAI | $0.20 / generation | Image-to-video |
| Gemini Omni Flash Video Edit | $0.25 / generation | Edit/restyle existing clips |
Rates as of 2026-09-24 — see the live pricing page for current numbers.
Quick start
POST https://genrelay.ai/v1/videos/generations
Authorization: Bearer YOUR_KEY
Content-Type: application/json
{
"model": "veo-3.1",
"prompt": "a product on a rotating turntable, studio lighting, 4K"
}
Video jobs are async — submit a generation request, then poll the job status endpoint until the output is ready. See the API docs for the full request/response schema.
Why one endpoint instead of managing each provider separately
- One API key and one prepaid balance across Google, ByteDance, and xAI video models — no separate billing relationships to manage.
- Switch models per request by changing one parameter, so you can route cheaper/faster models for drafts and premium models for final output.
- 1080p/4K output with native audio on supported models, managed from one console alongside your keys and usage.