Kling 3.0 Motion Control API

Complete API reference for Kling 3.0 Motion Control video generation by Kuaishou.

Model Variants

This API supports one model variant:

Model	ID	Description
Kling 3.0 Motion Control	`kling-3.0-motion-control`	Video generation using a text prompt, one reference image, and one reference video

Kling 3.0 Motion Control

Endpoint

POST /api/v1/video/task

Request Parameters

Required

Parameter	Type	Description
`model_name`	string	Must be `"kling-3.0-motion-control"`
`prompt`	string	Video generation prompt, up to 5000 characters
`images`	string[]	Exactly one reference image URL. Supports `.jpg`, `.jpeg`, and `.png`; file size must not exceed 50 MB. Width and height must each be at least 300 px, with an image aspect ratio between 1:2.5 and 2.5:1
`videos`	string[]	Exactly one reference video URL. Supports `.mp4` and `.mov`; file size must not exceed 100 MB. Width and height must each be between 340 and 3850 px, inclusive. Duration must be at least 3 seconds; the maximum depends on `character_orientation`
`character_orientation`	string	Required with no default. `"image"`: follows the character's orientation in the reference image; reference video must be 3-10 seconds. `"video"`: follows the character's orientation in the reference video; reference video must be 3-30 seconds. Both duration ranges include their endpoints
`resolution`	string	Output resolution: `"720p"` or `"1080p"`. Required with no default

Optional

Parameter	Type	Description
`keep_original_sound`	boolean	Whether to retain the reference video's original sound. Accepts `true` or `false`; defaults to `true`

Output duration follows the source video. The duration parameter is not supported. Output aspect ratio is determined by the input media, and aspect_ratio is not supported. The upstream provider has not specified which input determines the output aspect ratio or the exact selection rule.

Billing is based on the source video's duration rounded to the nearest whole second and the selected output resolution. For example, 5.4 seconds is billed as 5 seconds, and 5.5 seconds as 6 seconds.

Example Request

Replace the example media URLs with your own files that meet the requirements above.

javascript
const response = await fetch('https://api.flaq.ai/api/v1/video/task', {
  method: 'POST',
  headers: {
    'Authorization': 'Bearer YOUR_API_KEY',
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({
    model_name: 'kling-3.0-motion-control',
    prompt: 'Animate the character in the reference image using the dance movements from the reference video',
    images: ['https://example.com/character.jpg'],
    videos: ['https://example.com/dance-reference.mp4'],
    character_orientation: 'video',
    resolution: '720p',
    keep_original_sound: true
  })
});

const { data } = await response.json();
const taskId = data.task_id;

Response Format

Initial Response

Returns a task_id for polling:

json
{
  "code": 0,
  "message": "success",
  "data": {
    "task_id": "{task_id}",
    "response_url": "https://api.flaq.ai/api/v1/video/{task_id}",
    "task_status": "submitted"
  }
}

Poll with GET /api/v1/video/{task_id} to check status and retrieve the generated video.

Status Values

submitted: Task accepted and waiting to be processed
processing: Video generation in progress
succeed: Video generation completed
failed: Video generation failed

Best Practices

Reference Media: Supply exactly one image and one video, checking their formats, sizes, and dimensions before submitting
Orientation: Recheck the reference video's duration when switching character_orientation, especially when selecting "image"
Original Sound: Set keep_original_sound to false to disable retention of the reference video's original sound; omitting it retains the sound