Create video generation job

​

Veo image-to-video

​

Quick flow

1. Submit job → get task_id
2. Poll status → wait until succeeded for a public URL

​

Basics

• Endpoint: POST /v1/video/generations
• Auth: Bearer Token (Authorization: Bearer {{YOUR_API_KEY}})
• Content-Type: application/json

​

Request body

​

Image rules

• Formats: PNG, JPEG, JPG, WebP
• Size: ≤ 50MB file (POST body limit 20MB)
• Aspect: between 1:4 and 4:1

​

Response

{
  "id": "task_MQKsdsosLYRreob4Z4NzTFxZKtSoVRWZ",
  "task_id": "task_MQKsdsosLYRreob4Z4NzTFxZKtSoVRWZ",
  "object": "video",
  "model": "veo-2.0-generate-001",
  "status": "queued",
  "progress": 0,
  "created_at": 1775116001
}

​

Request example

{
  "model": "Veoq3-turbo",
  "prompt": "Contemporary dance, the people in the picture are performing contemporary dance.",
  "images": ["https://example.com/dance_start_frame.png"],
  "duration": 7
}

​

Notes

1. audio_type split (SFX vs speech) applies on Q2 / Q1 models only, per upstream docs.
2. Off-peak: Jobs not finished in 48h may auto-cancel with refund; you can cancel earlier.
3. is_rec: System prompt optimization when unsure (+ credits).
4. Callback signing: Verify Veo webhook signatures server-side.