Grok Imagine Video Series

POST

tasks

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>",
  "image_urls": [
    "<string>"
  ],
  "video_urls": [
    "<string>"
  ]
}
'

import requests

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

payload = {
    "model": "<string>",
    "action": "<string>",
    "prompt": "<string>",
    "duration": 123,
    "resolution": "<string>",
    "aspect_ratio": "<string>",
    "image_urls": ["<string>"],
    "video_urls": ["<string>"]
}
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>',
    image_urls: ['<string>'],
    video_urls: ['<string>']
  })
};

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  \"image_urls\": [\n    \"<string>\"\n  ],\n  \"video_urls\": [\n    \"<string>\"\n  ]\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  \"image_urls\": [\n    \"<string>\"\n  ],\n  \"video_urls\": [\n    \"<string>\"\n  ]\n}")
  .asString();

The xAI Grok Imagine video line offers three model IDs. The official grok-imagine-video-1.5 reaches 1080p, starts at 1 second and takes multiple reference images; the official grok-imagine-video generates at 480p / 720p and adds a video edit action, edit; the economy grok-imagine-video-1.5-rev offers 480p / 720p only, starts at 6 seconds, and takes up to 7 reference images with no extra fee. For images, see Grok Imagine Series.

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: grok-video-001" \
  -d '{
    "model": "grok-imagine-video-1.5",
    "action": "generate",
    "prompt": "A snowfield under the aurora, slow lateral pan, no text",
    "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" \
  -H "Idempotency-Key: grok-video-i2v-001" \
  -d '{
    "model": "grok-imagine-video-1.5",
    "action": "generate",
    "prompt": "The person in the first image wears the coat from the second, slowly turning toward the camera",
    "image_urls": [
      "https://cdn.example.com/portrait.jpg",
      "https://cdn.example.com/coat.jpg"
    ],
    "duration": 6,
    "resolution": "720p",
    "aspect_ratio": "9:16"
  }'

curl -X POST https://api.qingbo.ai/v1/tasks \
  -H "Authorization: Bearer $WAVE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "grok-imagine-video",
    "action": "edit",
    "prompt": "Restyle the footage as watercolor, keep the motion unchanged",
    "video_urls": ["https://cdn.example.com/source.mp4"]
  }'

curl -X POST https://api.qingbo.ai/v1/tasks \
  -H "Authorization: Bearer $WAVE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "grok-imagine-video-1.5-rev",
    "action": "generate",
    "prompt": "Keep the character consistent while running through a forest",
    "image_urls": [
      "https://cdn.example.com/ref-1.jpg",
      "https://cdn.example.com/ref-2.jpg",
      "https://cdn.example.com/ref-3.jpg"
    ],
    "duration": 6,
    "resolution": "720p",
    "aspect_ratio": "16:9"
  }'

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.

Request parameters

string

required

One of three IDs; see Available models.

string

default:"generate"

generate — text-to-video or image-to-video, on all three models; sending image_urls makes it image-to-video
edit — video edit, grok-imagine-video only; video_urls is required

string

required

Description of the video or the edit instruction, in English or Chinese.

integer

Output length in whole seconds; required under action: "generate". grok-imagine-video and grok-imagine-video-1.5 accept 1–15; grok-imagine-video-1.5-rev accepts 6–15. Under action: "edit", omit it or send 0; the output length follows the source video.

string

default:"480p"

Output resolution. grok-imagine-video-1.5 supports 480p, 720p and 1080p; the other two models support 480p and 720p. No effect under action: "edit", where the output keeps the source resolution (up to 720p).

string

Frame ratio; the accepted set differs by model, see Available models. grok-imagine-video-1.5 defaults to auto and needs an explicit ratio when reference images are sent; grok-imagine-video and grok-imagine-video-1.5-rev default to 16:9. No effect under action: "edit", where the source ratio is kept.

string[]

Reference image URLs, publicly reachable, for action: "generate" only. grok-imagine-video and grok-imagine-video-1.5 take multiple images; grok-imagine-video-1.5-rev takes up to 7.

string[]

Source video URLs, accepted only by action: "edit" on grok-imagine-video; one clip, up to 8.7 seconds.

See Submit Task for callback_url, callback_events, Prefer: wait, Idempotency-Key and the maximum-cost header. n is not supported — a task produces one result. seed, generate_audio and audio_urls are not supported. Any other parameter returns 400 and is not billed.

Limits

Model	Condition	Limit
`grok-imagine-video`	Every request	This model produces one result per task; n is not supported
`grok-imagine-video`	`action: "generate"`	generate does not take a source video; use the edit action to edit a video.
`grok-imagine-video`	`action: "edit"`	Video edit takes exactly one source video (up to 8.7 s) and no reference images. The output keeps the source length, aspect ratio and resolution (capped at 720p), so duration cannot be set (omit it or send 0) and resolution / aspect_ratio have no effect.
`grok-imagine-video-1.5`	Every request	The video endpoint does not accept n.
`grok-imagine-video-1.5`	Media inputs	This group does not accept video or audio inputs.
`grok-imagine-video-1.5`	Audio generation	The supplier does not document an audio-control parameter.
`grok-imagine-video-1.5`	`image_urls` provided	With reference images, set an explicit aspect ratio (1:1 / 16:9 / 9:16 / 4:3 / 3:4 / 3:2 / 2:3); auto cannot be used.
`grok-imagine-video-1.5-rev`	Every request	The video endpoint does not accept n.
`grok-imagine-video-1.5-rev`	Media inputs	This group does not accept video or audio inputs.
`grok-imagine-video-1.5-rev`	Audio generation	The supplier does not document an audio-control parameter.

Pricing

Billing is resolution tier × seconds; each resolution tier has its own per-second rate.

Model · action	Billing dimensions
`grok-imagine-video` · `generate`	`480p` / `720p` per-second rate × output seconds; reference images charged per image
`grok-imagine-video` · `edit`	Edit per-second rate × actual output seconds + source-video per-second rate × source seconds
`grok-imagine-video-1.5` · `generate`	`480p` / `720p` / `1080p` per-second rate × `duration`; reference images charged per image
`grok-imagine-video-1.5-rev` · `generate`	`480p` / `720p` per-second rate × `duration`; no fee for reference images

grok-imagine-video freezes quota at submission (edit freezes at the duration cap) and settles on the actual result when the task completes, refunding the difference.

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": "grok-imagine-video-1.5-rev",
  "action": "generate",
  "status": "completed",
  "progress": "100%",
  "created_at": 1775285160,
  "completed_at": 1775285352,
  "billing_status": "settled",
  "cost": 64530,
  "result": {
    "videos": [
      {
        "expires_at": 1775371560,
        "url": ["https://cdn.example.com/result.mp4"]
      }
    ]
  },
  "urls": {
    "get": "https://api.qingbo.ai/v1/tasks/task-wave1775285160b950328499",
    "cancel": "https://api.qingbo.ai/v1/tasks/task-wave1775285160b950328499/cancel"
  }
}

result.videos carries the generated video, one per task. Video links expire, so store them once retrieved. Full field reference: Query Task Status.

Available models

Model ID	Actions	Resolution	Duration (s)	Reference images
`grok-imagine-video`	`generate` `edit`	`480p` `720p`	1–15; `edit` follows the source	Multiple, charged per image
`grok-imagine-video-1.5`	`generate`	`480p` `720p` `1080p`	1–15	Multiple, charged per image
`grok-imagine-video-1.5-rev`	`generate`	`480p` `720p`	6–15	Up to 7, no extra fee

Aspect ratio sets:

grok-imagine-video-1.5: auto, 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3 (auto cannot be used with reference images)
grok-imagine-video: 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3
grok-imagine-video-1.5-rev: 16:9, 9:16, 1:1, 3:2, 2:3

Official vs reverse

The economy model shares the task endpoint and the parameter names with the official ones. Four things differ:

Resolution — 1080p exists only on grok-imagine-video-1.5; the economy model tops out at 720p.
Duration — the two official models start at 1 second, the economy model at 6 seconds; all cap at 15.
Reference images — both official models charge per image; grok-imagine-video-1.5-rev takes up to 7 with no extra fee.
Video edit — only grok-imagine-video supports action: "edit".

Grok Imagine Video 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>",
  "image_urls": [
    "<string>"
  ],
  "video_urls": [
    "<string>"
  ]
}
'

import requests

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

payload = {
    "model": "<string>",
    "action": "<string>",
    "prompt": "<string>",
    "duration": 123,
    "resolution": "<string>",
    "aspect_ratio": "<string>",
    "image_urls": ["<string>"],
    "video_urls": ["<string>"]
}
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>',
    image_urls: ['<string>'],
    video_urls: ['<string>']
  })
};

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  \"image_urls\": [\n    \"<string>\"\n  ],\n  \"video_urls\": [\n    \"<string>\"\n  ]\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  \"image_urls\": [\n    \"<string>\"\n  ],\n  \"video_urls\": [\n    \"<string>\"\n  ]\n}")
  .asString();

Text Generation

Image Generation

Video Generation

Audio

Tools & Embeddings

Task Management

Account & Usage

Grok Imagine Video Series

Quick start

Request parameters

Limits

Pricing

Response

Available models

Official vs reverse

​Quick start

​Request parameters

​Limits

​Pricing

​Response

​Available models

​Official vs reverse

​Related

Quick start

Request parameters

Limits

Pricing

Response

Available models

Official vs reverse

Related