Veo 3.1 Video API

Quick start

Breaking change (2025-12-10): generation_type → type; enum values now text-to-video / image-to-video / reference-to-video.
Platform: Wizzx gateway; Auth header Authorization: Bearer <your_api_key>.
Submit: POST https://api.wizzx.ai/api/v1/task/submit/veo3 → returns task_id.
Poll: POST https://api.wizzx.ai/api/v1/task/status with task_id.
Default model veo3_fast, resolution 720p, aspect ratio 16:9; veo3_lite is available for lower-cost generation and veo3_quality for the highest fidelity.

Auth

Every request:

Authorization: Bearer <your_api_key>

Wizzx gateway validates and forwards; no extra params needed.

Submit job

Endpoint: POST https://api.wizzx.ai/api/v1/task/submit/veo3, Content-Type: application/json

Required

Field	Type	Notes
`prompt`	string	Video description
`type`	string	`text-to-video` / `image-to-video` / `reference-to-video`

Optional

Field	Type	Default	Notes
`image_urls`	[]string	-	1-3 images (see rules below)
`duration`	int	`8`	`4` / `6` / `8` seconds; `veo3_quality` accepts `6` or `8` only
`resolution`	string	`720p`	`720p` / `1080p` / `4k`
`model`	string	`veo3_fast`	`veo3_fast` / `veo3_lite` / `veo3_quality`
`aspect_ratio`	string	`16:9`	`16:9` / `9:16` / `Auto`
`seeds`	int	-	10000-99999
`enable_translation`	bool	-	Optional provider translation hint
`watermark`	string	-	Custom watermark text

Type rules

type	Description	`image_urls` requirement
`text-to-video`	Pure text to video	Omit `image_urls`
`image-to-video`	First frame / first+last	1-2 images (1 = first frame, 2 = first+last)
`reference-to-video`	Reference images	1-3 images; Fast and Lite only (Quality is rejected); duration must be 8 seconds

type is always required. reference-to-video requires duration=8 and veo3_fast or veo3_lite. veo3_quality accepts 6 or 8 seconds only; 4 seconds returns HTTP 422.

Request: Text-to-Video

curl -X POST https://api.wizzx.ai/api/v1/task/submit/veo3 \
  -H "Authorization: Bearer <api_key>" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "A dog running through a neon city at night",
    "type": "text-to-video",
    "aspect_ratio": "16:9",
    "model": "veo3_lite",
    "resolution": "720p",
    "seeds": 12345
  }'

Request: Image-to-Video (first/last frames)

curl -X POST https://api.wizzx.ai/api/v1/task/submit/veo3 \
  -H "Authorization: Bearer <api_key>" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "Make the scene dynamic while keeping the character consistent",
    "type": "image-to-video",
    "image_urls": [
      "https://example.com/frame1.jpg",
      "https://example.com/frame2.jpg"
    ],
    "aspect_ratio": "9:16"
  }'

Request: Text-to-Video with Veo 3.1 Quality

curl -X POST https://api.wizzx.ai/api/v1/task/submit/veo3 \
  -H "Authorization: Bearer <api_key>" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "A slow dolly shot through a rain-soaked night market",
    "type": "text-to-video",
    "model": "veo3_quality",
    "resolution": "1080p",
    "duration": 8
  }'

Request: Reference-to-Video with Veo 3.1 Lite

curl -X POST https://api.wizzx.ai/api/v1/task/submit/veo3 \
  -H "Authorization: Bearer <api_key>" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "Product turntable style, soft studio light",
    "type": "reference-to-video",
    "image_urls": [
      "https://example.com/ref1.jpg",
      "https://example.com/ref2.jpg",
      "https://example.com/ref3.jpg"
    ],
    "model": "veo3_lite",
    "resolution": "1080p",
    "aspect_ratio": "9:16",
    "duration": 8,
    "enable_translation": true
  }'

Success response:

{
  "code": 0,
  "message": "success",
  "data": {
    "task_id": "4bbbc9e0-8de8-4f35-a31e-8596c1a6d18a",
    "model_key": "veo3",
    "credits_deducted": 210
  },
  "request_id": "d54c66a8-0f0f-439c-bd3a-36f3fd275881"
}

Job status

Endpoint: POST https://api.wizzx.ai/api/v1/task/status Body:

{
  "task_id": "4bbbc9e0-8de8-4f35-a31e-8596c1a6d18a"
}

Response:

{
  "code": 0,
  "message": "success",
  "data": {
    "task_id": "4bbbc9e0-8de8-4f35-a31e-8596c1a6d18a",
    "status": "SUCCEEDED",
    "result_urls": [
      "https://cdn.example.com/video.mp4"
    ],
    "error": "",
    "credits_deducted": 210
  },
  "request_id": "499347b1-ff99-4ba2-a977-ad655431919d"
}

Client-visible states: PENDING, QUEUED, PROCESSING, REVIEW_REQUIRED, SUCCEEDED, and FAILED.

Pricing

Veo 3.1 uses a fixed per-video charge selected by model and resolution; duration does not change the price (Fast and Lite: 4, 6, or 8 seconds; Quality: 6 or 8 seconds).

Model	720p	1080p	4K
`veo3_fast`	360	390	1080
`veo3_lite`	180	210	900
`veo3_quality`	1924	1924	2885

Read /pricing before submission for the current tier. Each video is charged a fixed price by model variant and resolution at 1000 credits = $1.

FAQ

Where did generation_type go? Unified to type, enum now lower-case with hyphens.
Need callback URL? No, poll task/status.
How to reproduce results? Provide seeds (10000-99999).
Non-English prompt? Set enable_translation=true explicitly if provider translation is wanted.
Difference between 1 vs 2 images in image-to-video? One image = first frame; two images = first + last frame with auto transition.
Does duration multiply the Veo charge? No. Veo is charged a fixed price per video; model variant and resolution select the charge.

Sources

Google DeepMind: Veo

Getting started

Agent integrations

Video Generation

Image Generation

Audio Generation

Veo 3.1 Video API

Quick start

Auth

Submit job

Required

Optional

Type rules

Request: Text-to-Video

Request: Image-to-Video (first/last frames)

Request: Text-to-Video with Veo 3.1 Quality

Request: Reference-to-Video with Veo 3.1 Lite

Job status

Pricing

FAQ

Sources

​Quick start

​Auth

​Submit job

​Required

​Optional

​Type rules

​Request: Text-to-Video

​Request: Image-to-Video (first/last frames)

​Request: Text-to-Video with Veo 3.1 Quality

​Request: Reference-to-Video with Veo 3.1 Lite

​Job status

​Pricing

​FAQ

​Sources

Quick start

Auth

Submit job

Required

Optional

Type rules

Request: Text-to-Video

Request: Image-to-Video (first/last frames)

Request: Text-to-Video with Veo 3.1 Quality

Request: Reference-to-Video with Veo 3.1 Lite

Job status

Pricing

FAQ

Sources