Veo Series

POST

tasks

Veo Series

curl --request POST \
  --url https://api.qingbo.ai/v1/tasks \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "<string>",
  "action": "<string>",
  "prompt": "<string>",
  "duration": 123,
  "resolution": "<string>",
  "aspect_ratio": "<string>",
  "first_frame_image": "<string>",
  "last_frame_image": "<string>",
  "negative_prompt": "<string>",
  "seed": 123,
  "generate_audio": true,
  "person_generation": "<string>",
  "resize_mode": "<string>",
  "image_urls": [
    "<string>"
  ],
  "generation_type": "<string>",
  "enable_gif": true,
  "ref_task_id": "<string>",
  "raw": true
}
'

import requests

url = "https://api.qingbo.ai/v1/tasks"

payload = {
    "model": "<string>",
    "action": "<string>",
    "prompt": "<string>",
    "duration": 123,
    "resolution": "<string>",
    "aspect_ratio": "<string>",
    "first_frame_image": "<string>",
    "last_frame_image": "<string>",
    "negative_prompt": "<string>",
    "seed": 123,
    "generate_audio": True,
    "person_generation": "<string>",
    "resize_mode": "<string>",
    "image_urls": ["<string>"],
    "generation_type": "<string>",
    "enable_gif": True,
    "ref_task_id": "<string>",
    "raw": True
}
headers = {"Content-Type": "application/json"}

response = requests.post(url, json=payload, headers=headers)

print(response.text)

const options = {
  method: 'POST',
  headers: {'Content-Type': 'application/json'},
  body: JSON.stringify({
    model: '<string>',
    action: '<string>',
    prompt: '<string>',
    duration: 123,
    resolution: '<string>',
    aspect_ratio: '<string>',
    first_frame_image: '<string>',
    last_frame_image: '<string>',
    negative_prompt: '<string>',
    seed: 123,
    generate_audio: true,
    person_generation: '<string>',
    resize_mode: '<string>',
    image_urls: ['<string>'],
    generation_type: '<string>',
    enable_gif: true,
    ref_task_id: '<string>',
    raw: true
  })
};

fetch('https://api.qingbo.ai/v1/tasks', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

package main

import (
	"fmt"
	"strings"
	"net/http"
	"io"
)

func main() {

	url := "https://api.qingbo.ai/v1/tasks"

	payload := strings.NewReader("{\n  \"model\": \"<string>\",\n  \"action\": \"<string>\",\n  \"prompt\": \"<string>\",\n  \"duration\": 123,\n  \"resolution\": \"<string>\",\n  \"aspect_ratio\": \"<string>\",\n  \"first_frame_image\": \"<string>\",\n  \"last_frame_image\": \"<string>\",\n  \"negative_prompt\": \"<string>\",\n  \"seed\": 123,\n  \"generate_audio\": true,\n  \"person_generation\": \"<string>\",\n  \"resize_mode\": \"<string>\",\n  \"image_urls\": [\n    \"<string>\"\n  ],\n  \"generation_type\": \"<string>\",\n  \"enable_gif\": true,\n  \"ref_task_id\": \"<string>\",\n  \"raw\": true\n}")

	req, _ := http.NewRequest("POST", url, payload)

	req.Header.Add("Content-Type", "application/json")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}

HttpResponse<String> response = Unirest.post("https://api.qingbo.ai/v1/tasks")
  .header("Content-Type", "application/json")
  .body("{\n  \"model\": \"<string>\",\n  \"action\": \"<string>\",\n  \"prompt\": \"<string>\",\n  \"duration\": 123,\n  \"resolution\": \"<string>\",\n  \"aspect_ratio\": \"<string>\",\n  \"first_frame_image\": \"<string>\",\n  \"last_frame_image\": \"<string>\",\n  \"negative_prompt\": \"<string>\",\n  \"seed\": 123,\n  \"generate_audio\": true,\n  \"person_generation\": \"<string>\",\n  \"resize_mode\": \"<string>\",\n  \"image_urls\": [\n    \"<string>\"\n  ],\n  \"generation_type\": \"<string>\",\n  \"enable_gif\": true,\n  \"ref_task_id\": \"<string>\",\n  \"raw\": true\n}")
  .asString();

Google’s Veo 3.1 video line comes in two routes. The official route, veo3.1-fast and veo3.1-quality, offers 4 / 6 / 8 second output, first and last frames, native audio and seed, billed per second. The economy route, veo3.1-lite, veo3.1-fast-rev and veo3.1-quality-rev, reaches Veo through an aggregator, always outputs eight seconds and is billed per finished clip; the two -rev tiers can also extend a video they generated earlier. veo3.1-lite (Veo 3.1 Lite Economy) belongs to the economy route and is not Google’s official Veo 3.1 Lite. Both routes offer 720p, 1080p and 4k; they differ in both parameters and billing, and never switch to each other.

Quick start

curl -X POST https://api.qingbo.ai/v1/tasks \
  -H "Authorization: Bearer $WAVE_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: veo-demo-001" \
  -d '{
    "model": "veo3.1-fast",
    "action": "generate",
    "prompt": "Aerial sunrise over snow peaks, clouds rolling, slow push in",
    "duration": 8,
    "resolution": "1080p",
    "aspect_ratio": "16:9"
  }'

curl -X POST https://api.qingbo.ai/v1/tasks \
  -H "Authorization: Bearer $WAVE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "veo3.1-quality",
    "action": "generate",
    "prompt": "The subject looks up and smiles as the light warms",
    "first_frame_image": "https://cdn.example.com/first.jpg",
    "last_frame_image": "https://cdn.example.com/last.jpg",
    "duration": 8,
    "resolution": "1080p",
    "resize_mode": "crop"
  }'

curl -X POST https://api.qingbo.ai/v1/tasks \
  -H "Authorization: Bearer $WAVE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "veo3.1-quality",
    "action": "generate",
    "prompt": "Waves break against the rocks, gulls calling, a distant ship horn",
    "duration": 8,
    "resolution": "720p",
    "aspect_ratio": "16:9",
    "generate_audio": true,
    "seed": 42
  }'

curl -X POST https://api.qingbo.ai/v1/tasks \
  -H "Authorization: Bearer $WAVE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "veo3.1-lite",
    "action": "generate",
    "prompt": "A seaside town at sunset, pigeons crossing the clock tower",
    "duration": 8,
    "resolution": "720p",
    "aspect_ratio": "16:9"
  }'

curl -X POST https://api.qingbo.ai/v1/tasks \
  -H "Authorization: Bearer $WAVE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "veo3.1-quality-rev",
    "action": "generate",
    "prompt": "The camera pushes from the corner to the shop door as the neon comes on",
    "generation_type": "frame",
    "image_urls": [
      "https://cdn.example.com/first.jpg",
      "https://cdn.example.com/last.jpg"
    ],
    "duration": 8,
    "resolution": "1080p"
  }'

curl -X POST https://api.qingbo.ai/v1/tasks \
  -H "Authorization: Bearer $WAVE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "veo3.1-fast-rev",
    "action": "generate",
    "prompt": "Put the character, the outfit and the location from these images in one shot",
    "generation_type": "reference",
    "image_urls": [
      "https://cdn.example.com/character.jpg",
      "https://cdn.example.com/outfit.jpg",
      "https://cdn.example.com/scene.jpg"
    ],
    "duration": 8,
    "resolution": "720p"
  }'

curl -X POST https://api.qingbo.ai/v1/tasks \
  -H "Authorization: Bearer $WAVE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "veo3.1-fast-rev",
    "action": "extend",
    "ref_task_id": "task-wave1775285160b950328499",
    "prompt": "The camera keeps moving forward, into the shop",
    "resolution": "720p",
    "raw": false
  }'

A successful submission returns a task_id. Retrieve the result with GET /v1/tasks/{task_id}, wait inline with Prefer: wait, or configure a webhook. Chaining an extension takes two steps. Submit action: "generate" on veo3.1-fast-rev or veo3.1-quality-rev and keep the task_id; once that task is completed, pass its task_id as ref_task_id on an action: "extend" request. ref_task_id is the task_id returned by that first task, not an upstream ID, and it must be a completed generate task on the same model under the same API key. An extension is 720p by default; a 4k extension needs a 4k parent.

Request parameters

Both routes

string

required

Model ID; see Available models.

string

default:"generate"

generate — generate a video; supported by all five models
extend — continue a finished video; only veo3.1-fast-rev and veo3.1-quality-rev support it, and ref_task_id is required

string

required

Video description. Under action: "extend" it describes the continuation.

integer

Output duration in seconds; required under action: "generate". The official route takes 4, 6 or 8, and resolution 1080p or 4k takes 8 only; the economy route is fixed at 8. action: "extend" does not take this parameter — the continuation’s length is decided upstream.

string

default:"720p"

Output resolution: 720p, 1080p or 4k. Also accepted under action: "extend", where it defaults to 720p; a 4k extension needs a 4k parent.

string

default:"16:9"

Frame ratio: 16:9 or 9:16.

Official route (`veo3.1-fast` / `veo3.1-quality`)

string

First-frame image URL.

string

Last-frame image URL; must be sent together with first_frame_image.

string

Negative prompt describing what to keep out of the video.

integer

Random seed, for reproducing a result.

boolean

default:"false"

Generate native audio. Enabling it moves the request to the audio pricing tier.

string

Person policy: allow_adult allows adult people / disallow generates none.

string

How to fit a frame image whose ratio differs from the output: pad or crop.

Economy route (`veo3.1-lite` / `veo3.1-fast-rev` / `veo3.1-quality-rev`)

string[]

Reference image URLs, supported by veo3.1-fast-rev and veo3.1-quality-rev; how many depends on generation_type. veo3.1-lite is text-to-video only and takes no images.

string

How the images are used; required whenever image_urls is present:

frame — treat them as first / last frames; 1-2 images
reference — treat them as reference material; 1-3 images, veo3.1-fast-rev only

boolean

default:"false"

Also output a GIF. Cannot be combined with resolution 1080p or 4k.

string

required

The parent task being extended; required under action: "extend". Use the task_id returned by the generate task.

boolean

default:"false"

Under action: "extend", whether to return only the newly generated segment. false returns the joined clip.

See Submit a task for callback_url, callback_events, Prefer: wait, Idempotency-Key, and the maximum-cost header. The official route does not support image_urls, generation_type, enable_gif, ref_task_id or raw; the economy route does not support first_frame_image, last_frame_image, negative_prompt, seed, generate_audio, person_generation or resize_mode. Any other parameter returns 400 and is not billed.

Limits

Models	Condition	Limit
All	`action: "generate"`	Generation requires prompt and duration.
`veo3.1-fast`, `veo3.1-quality`	`last_frame_image` provided	A last frame requires a first frame.
`veo3.1-fast`, `veo3.1-quality`	`resolution` is `1080p` or `4k`	1080p and 4K require an 8 second duration.
`veo3.1-fast-rev`, `veo3.1-quality-rev`	`action: "generate"`	Reverse generation is fixed at eight seconds.
`veo3.1-fast-rev`, `veo3.1-quality-rev`	`action: "extend"`	Extend accepts a parent task, prompt, raw and resolution (default 720p) only; a 4K extension needs a 4K parent.
`veo3.1-fast-rev`, `veo3.1-quality-rev`	`image_urls` provided	generation_type is required with images.
`veo3.1-fast-rev`, `veo3.1-quality-rev`	`generation_type: "frame"`	Frame mode accepts one or two images.
`veo3.1-fast-rev`	`generation_type: "reference"`	Reference mode accepts up to three images.
`veo3.1-lite`, `veo3.1-fast-rev`, `veo3.1-quality-rev`	`enable_gif: true` with `resolution` `1080p` or `4k`	GIF output cannot be combined with 1080p or 4K.

Item	Models	Maximum
Reference images	`veo3.1-fast-rev`	3
Reference images	`veo3.1-quality-rev`	2
Reference images	`veo3.1-lite`	No images
Frame images	`veo3.1-fast`, `veo3.1-quality`	1 first frame + 1 last frame
Duration	`veo3.1-fast`, `veo3.1-quality`	4 / 6 / 8 seconds, 8 at most; `1080p` and `4k` are 8 only
Duration	`veo3.1-lite`, `veo3.1-fast-rev`, `veo3.1-quality-rev`	Fixed at 8 seconds

The ref_task_id on action: "extend" must point at a completed generate task on the same model. Pointing at another model, another action or an unfinished task returns 400 and is not billed.

Pricing

Official route

veo3.1-fast and veo3.1-quality are billed resolution tier × seconds: resolution picks the tier (720p, 1080p or 4k), and the seconds come from duration (4, 6 or 8; 1080p and 4k take 8 only). generate_audio: true switches to the audio tier’s per-second rate, still multiplied by duration alone. All three resolutions have an audio tier. Frame images, negative_prompt and seed do not affect the price.

Economy route

veo3.1-lite, veo3.1-fast-rev and veo3.1-quality-rev are billed per finished clip: one charge per generation, independent of length (output is always eight seconds). Resolution is the only tiering dimension; the number of reference images and the generation_type do not affect the price. action: "extend" is also one charge per finished clip, independent of how long the continuation is, tiered by the resolution in the request (720p when omitted); raw does not affect the price.

Per-model rates are in price_config on GET /v1/models, and in the console’s Model Market.What a single call actually cost is the cost field on the task response — an integer quota at 500,000 quota = 1 USD.

You are only billed for a successfully generated video. Failed and cancelled tasks, and tasks that return no usable video, are refunded in full. The final charge is the cost field on the task response — an integer quota, not dollars.

Response

Completed task

{
  "task_id": "task-wave1775285160b950328499",
  "model": "veo3.1-fast",
  "action": "generate",
  "status": "completed",
  "progress": "100%",
  "created_at": 1775285160,
  "completed_at": 1775285412,
  "result": {
    "videos": [
      {
        "url": ["https://cdn.example.com/result.mp4"],
        "expires_at": 1775621160
      }
    ]
  },
  "urls": {
    "get": "https://api.qingbo.ai/v1/tasks/task-wave1775285160b950328499",
    "cancel": "https://api.qingbo.ai/v1/tasks/task-wave1775285160b950328499/cancel"
  },
  "billing_status": "settled",
  "cost": 288000
}

Generated videos are returned in result.videos; expires_at is when the link stops working, so store the file before then. See Query Task Status for the full field list.

Available models

Official route — billed per second, with frame control and native audio.

Model ID	Duration	Resolution	Image input	Native audio
`veo3.1-fast`	4 / 6 / 8s	`720p` `1080p` `4k` (1080p and 4k are 8s only)	First + last frame	Yes
`veo3.1-quality`	4 / 6 / 8s	`720p` `1080p` `4k` (1080p and 4k are 8s only)	First + last frame	Yes

Economy route — one price per finished clip, always eight seconds, no native audio.

Model ID	Actions	Resolution	Image input	GIF
`veo3.1-lite`	`generate`	`720p` `1080p` `4k`	None	`enable_gif` (720p only)
`veo3.1-fast-rev`	`generate` `extend`	`720p` `1080p` `4k`	`frame` 1-2 / `reference` 1-3	`enable_gif` (720p only)
`veo3.1-quality-rev`	`generate` `extend`	`720p` `1080p` `4k`	`frame` 1-2	`enable_gif` (720p only)

Veo Series

curl --request POST \
  --url https://api.qingbo.ai/v1/tasks \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "<string>",
  "action": "<string>",
  "prompt": "<string>",
  "duration": 123,
  "resolution": "<string>",
  "aspect_ratio": "<string>",
  "first_frame_image": "<string>",
  "last_frame_image": "<string>",
  "negative_prompt": "<string>",
  "seed": 123,
  "generate_audio": true,
  "person_generation": "<string>",
  "resize_mode": "<string>",
  "image_urls": [
    "<string>"
  ],
  "generation_type": "<string>",
  "enable_gif": true,
  "ref_task_id": "<string>",
  "raw": true
}
'

import requests

url = "https://api.qingbo.ai/v1/tasks"

payload = {
    "model": "<string>",
    "action": "<string>",
    "prompt": "<string>",
    "duration": 123,
    "resolution": "<string>",
    "aspect_ratio": "<string>",
    "first_frame_image": "<string>",
    "last_frame_image": "<string>",
    "negative_prompt": "<string>",
    "seed": 123,
    "generate_audio": True,
    "person_generation": "<string>",
    "resize_mode": "<string>",
    "image_urls": ["<string>"],
    "generation_type": "<string>",
    "enable_gif": True,
    "ref_task_id": "<string>",
    "raw": True
}
headers = {"Content-Type": "application/json"}

response = requests.post(url, json=payload, headers=headers)

print(response.text)

const options = {
  method: 'POST',
  headers: {'Content-Type': 'application/json'},
  body: JSON.stringify({
    model: '<string>',
    action: '<string>',
    prompt: '<string>',
    duration: 123,
    resolution: '<string>',
    aspect_ratio: '<string>',
    first_frame_image: '<string>',
    last_frame_image: '<string>',
    negative_prompt: '<string>',
    seed: 123,
    generate_audio: true,
    person_generation: '<string>',
    resize_mode: '<string>',
    image_urls: ['<string>'],
    generation_type: '<string>',
    enable_gif: true,
    ref_task_id: '<string>',
    raw: true
  })
};

fetch('https://api.qingbo.ai/v1/tasks', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

package main

import (
	"fmt"
	"strings"
	"net/http"
	"io"
)

func main() {

	url := "https://api.qingbo.ai/v1/tasks"

	payload := strings.NewReader("{\n  \"model\": \"<string>\",\n  \"action\": \"<string>\",\n  \"prompt\": \"<string>\",\n  \"duration\": 123,\n  \"resolution\": \"<string>\",\n  \"aspect_ratio\": \"<string>\",\n  \"first_frame_image\": \"<string>\",\n  \"last_frame_image\": \"<string>\",\n  \"negative_prompt\": \"<string>\",\n  \"seed\": 123,\n  \"generate_audio\": true,\n  \"person_generation\": \"<string>\",\n  \"resize_mode\": \"<string>\",\n  \"image_urls\": [\n    \"<string>\"\n  ],\n  \"generation_type\": \"<string>\",\n  \"enable_gif\": true,\n  \"ref_task_id\": \"<string>\",\n  \"raw\": true\n}")

	req, _ := http.NewRequest("POST", url, payload)

	req.Header.Add("Content-Type", "application/json")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}

HttpResponse<String> response = Unirest.post("https://api.qingbo.ai/v1/tasks")
  .header("Content-Type", "application/json")
  .body("{\n  \"model\": \"<string>\",\n  \"action\": \"<string>\",\n  \"prompt\": \"<string>\",\n  \"duration\": 123,\n  \"resolution\": \"<string>\",\n  \"aspect_ratio\": \"<string>\",\n  \"first_frame_image\": \"<string>\",\n  \"last_frame_image\": \"<string>\",\n  \"negative_prompt\": \"<string>\",\n  \"seed\": 123,\n  \"generate_audio\": true,\n  \"person_generation\": \"<string>\",\n  \"resize_mode\": \"<string>\",\n  \"image_urls\": [\n    \"<string>\"\n  ],\n  \"generation_type\": \"<string>\",\n  \"enable_gif\": true,\n  \"ref_task_id\": \"<string>\",\n  \"raw\": true\n}")
  .asString();

Text Generation

Image Generation

Video Generation

Audio

Tools & Embeddings

Task Management

Account & Usage

Quick start

Request parameters

Both routes

Official route (`veo3.1-fast` / `veo3.1-quality`)

Economy route (`veo3.1-lite` / `veo3.1-fast-rev` / `veo3.1-quality-rev`)

Limits

Pricing

Official route

Economy route

Response

Available models

​Quick start

​Request parameters

​Both routes

​Official route (veo3.1-fast / veo3.1-quality)

​Economy route (veo3.1-lite / veo3.1-fast-rev / veo3.1-quality-rev)

​Limits

​Pricing

​Official route

​Economy route

​Response

​Available models

​Related

Quick start

Request parameters

Both routes

Official route (`veo3.1-fast` / `veo3.1-quality`)

Economy route (`veo3.1-lite` / `veo3.1-fast-rev` / `veo3.1-quality-rev`)

Limits

Pricing

Official route

Economy route

Response

Available models

Related