Create a video generation task
Endpoint: POST https://4zapi.com/v1/video/create
Unified video interface supports VEO, Sora-2, Grok-Video and other models.
Distinguish specific models through the model field; return id after submission and poll through GET /v1/video/query.
Request body (application/json)
model(string, required): Model name ["veo3", "veo3-fast", "veo3-pro", "sora-2", "grok-imagine-video-1.5"]prompt(string, required): Video content descriptionaspect_ratio(string): Aspect ratio (VEO series) ["16:9", "9:16", "1:1"]duration: Video duration (seconds)enable_upsample(boolean): Whether to enable upsampling (VEO)enhance_prompt(boolean): Whether to automatically optimize prompt words (VEO)images(array<string>): Reference image URL list (Tusheng Video)orientation(string): Direction (Sora-2) ["portrait", "landscape"]seconds: Duration compatible fieldsize(string): Clarity (Sora-2) ["large", "small"]type: Omni series task typeswatermark(boolean): Whether to add watermark; false=do not add
Request example
{
"aspect_ratio": "16:9",
"enhance_prompt": true,
"model": "veo3",
"prompt": "A majestic eagle soaring over mountains at sunset"
}