AI Video

POSThttps://sangtao.ai/api/v2/videos/generate

Reference for AI Video.

Generate AI videos from text prompts or images. Supports text-to-video and image-to-video modes. Returns a jobId for polling.

Request Body

Field	Type	Description
model*	string	Model slug from GET /api/v2/models?category=Video
prompt*	string	Text description of the video to generate
imageId	string	Source image job ID for image-to-video (from a completed image generation job)
imageIds	string[]	Multiple source image job IDs (for models supporting multi-image input)
duration	int	Video duration in seconds. Depends on model (5 or 10s typically)
resolution	string	Output resolution: "720p", "1080p". Default varies by model
aspectRatio	string	"16:9" \| "9:16" \| "1:1". Default: "16:9"
mode	string	Generation mode (model-specific). E.g. "standard", "pro"
modeImage	string	Image processing mode for image-to-video. Model-specific option

Tip: Use GET /api/v2/models?category=Video to list available video models and their supported options.

cURL

bash

curl -X POST https://sangtao.ai/api/v2/videos/generate \
  -H "X-Api-Key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "veo-3.1-fast",
    "prompt": "A drone shot flying over rice terraces in Sapa, golden hour lighting",
    "aspectRatio": "16:9",
    "duration": 5
  }'

Text-to-Video Example

Generate a video from a text prompt only:

json

{
  "model": "veo-3.1-fast",
  "prompt": "A drone shot flying over rice terraces in Sapa, golden hour lighting",
  "aspectRatio": "16:9",
  "duration": 5
}

Image-to-Video Example

Animate a generated image into a video. First generate an image, then use its jobId as imageId:

json

{
  "model": "veo-3.1-fast",
  "prompt": "Camera slowly zooms in, gentle wind blowing",
  "imageId": "img-abc123",
  "duration": 5
}

Response

json

{
  "success": true,
  "data": {
    "jobId": "vid-xyz789"
  }
}

Job Polling

Poll GET /api/v2/jobs/{jobId} every 5 seconds. Video generation typically takes 30s-3min.

json

// GET /api/v2/jobs/{jobId}
{
  "success": true,
  "data": {
    "status": "complete",
    "resultUrl": "https://cdn.sangtao.ai/videos/result.mp4",
    "thumbnailUrl": "https://cdn.sangtao.ai/videos/thumb.jpg"
  }
}

Credit Flow

Phase	Description
Lock	Credits are estimated and locked when you submit the job
Calculate	After processing, actual cost is calculated based on scenes and voice usage
Settle	When complete, excess credits are automatically refunded

Note: lockedCreditCost in the response is a high estimate. Actual cost is usually lower.

Error Codes

Code	When
400	Invalid input (story too short/long, model/voice not found...)
401	Unauthorized — missing or invalid API key / token
402	Insufficient credits or premium voice requires top-up
429	Already 4 jobs running concurrently
503	System is processing too many requests. Please retry after 30 seconds

PreviousUpload directly NextRender Video