kwaivgi/kling-video/v3/pro/motion-control

Name: Kling Video V3 Pro
Author: KwaiVGI

Kling Video V3 Pro is KwaiVGI's video-to-video AI model. Transform, enhance, and edit video content using text prompts - from style changes to object manipulation and scene modification.

Input

Prompt(string)

Image URL or upload(string)

Upload

PNG, JPEG, WebP, or GIF · 20 MiB maximum

Reference image URL. The characters, backgrounds, and other elements in the generated video are based on this reference image. Characters should have clear body proportions, avoid occlusion, and occupy more than 5% of the image area.

This field is required.

Video Url(string)

Reference video URL. The character actions in the generated video will be consistent with this reference video. Should contain a realistic style character with entire body or upper body visible, including head, without obstruction. Duration limit depends on character_orientation: 10s max for 'image', 30s max for 'video'.

Keep Original Sound(boolean)

Whether to keep the original sound from the reference video.

Character Orientation(string)

Controls whether the output character's orientation matches the reference image or video. 'video': orientation matches reference video - better for complex motions (max 30s). 'image': orientation matches reference image - better for following camera movements (max 10s). Allowed values: image, video.

Idle

Example output — click Run to generate your own

Production endpoint

Send your first request

OpenAI-compatible endpoint with unified authentication and usage tracking.

POSThttps://api.sandbase.ai/v1/run

Model IDkwaivgi/kling-video/v3/pro/motion-control

Input Schema

5 parameters · 2 required · 3 optional

Parameter	Type	Required	Description
`image`	string	Required	Reference image URL. The characters, backgrounds, and other elements in the generated video are based on this reference image. Characters should have clear body proportions, avoid occlusion, and occupy more than 5% of the image area.
`prompt`	string	Required	Max length: 2500
`video`	string	Optional	Reference video URL. The character actions in the generated video will be consistent with this reference video. Should contain a realistic style character with entire body or upper body visible, including head, without obstruction. Duration limit depends on character_orientation: 10s max for 'image', 30s max for 'video'.
`keep_original_sound`	boolean	Optional	Whether to keep the original sound from the reference video. · Default: true
`character_orientation`	string	Optional	Controls whether the output character's orientation matches the reference image or video. 'video': orientation matches reference video - better for complex motions (max 30s). 'image': orientation matches reference image - better for following camera movements (max 10s). · Options: image, video imagevideo

Output Schema

Field	Type	Description
`id`	string	Unique identifier for the generation task
`status`	string	Task status: pending, running, completed, failed, timeout
`model`	string	Model used for the generation
`outputs`	array	Array of output items
`outputs[].url`	string	URL of the generated artifact
`outputs[].content_type`	string	MIME type (e.g. image/png, video/mp4)
`error`	object \| null	Error details if failed, null on success
`error.type`	string	Machine-readable error type code
`error.message`	string	Human-readable error description

Async Workflow

This model uses asynchronous execution. Submit a request and poll for the result.

Submit — POST to /v1/run, receive an id
Poll — GET /v1/run/{id} until status is completed, failed, or timeout
Retrieve — Read outputs from the completed response

Code Examples

Ready-to-run snippets

# Step 1: Submit
curl -X POST https://api.sandbase.ai/v1/run \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
  "model": "kwaivgi/kling-video/v3/pro/motion-control",
  "image": "https://static.sandbase.ai/examples/mirrored/f77b827ca02a-TNErq9yD7ZxGRATjfAqnh_EIgJSN67.png",
  "video": "https://static.sandbase.ai/examples/mirrored/f1ecbb471a16-hklvF__w53diz6Rve7f5__JuDW2xl0mr6sJ_Kjz3Vxe_vidoeook--1-_1.mp4",
  "prompt": "A man dancing",
  "keep_original_sound": true,
  "character_orientation": "image"
}'

# Step 2: Poll result (replace <id>)
curl https://api.sandbase.ai/v1/run/<id> \
  -H "Authorization: Bearer YOUR_API_KEY"

← Back to Kling Video V3 Pro

API README

Kling Video V3 Pro

kwaivgi/kling-video/v3/pro/motion-control is centered on kling Video V3 Pro is KwaiVGI's video-to-video AI model. The capability is useful when that specific transformation is the main objective, because the request can be planned around the documented source material and intended result rather than a generic generation workflow.

It gives teams a concrete way to decide whether the endpoint fits the available asset and desired change while keeping technical parameter details in the specifications below.

Highlights

Model capability. Within the kling-video/v3/pro/motion-control endpoint, kwaivgi/kling-video/v3/pro/motion-control documents edit video content using text prompts - from style changes to object manipulation and scene modification.
Core workflow. Within the kling-video/v3/pro/motion-control endpoint, kwaivgi/kling-video/v3/pro/motion-control documents video to video.
Creative control. Within the kling-video/v3/pro/motion-control endpoint, kwaivgi/kling-video/v3/pro/motion-control documents controls whether the output character's orientation matches the reference image or video. 'video': orientation matches reference video - better for complex motions (max 30s). 'image': orientation matches reference image - better for following camera movements (max 10s).
Source guidance. Within the kling-video/v3/pro/motion-control endpoint, kwaivgi/kling-video/v3/pro/motion-control documents reference image URL. The characters, backgrounds, and other elements in the generated video are based on this reference image. Characters should have clear body proportions, avoid occlusion, and occupy more than 5% of the image area.

Pricing

Local price formula: (usage.duration ?? (media_length(params.video ?? 0) > 0 ? media_length(params.video) : 5)) * 0.168

Technical Specs

Model ID: kwaivgi/kling-video/v3/pro/motion-control
Request fields: character_orientation (string, optional, enum: image/video); image (string, required); keep_original_sound (boolean, optional); prompt (string, required); video (string, optional)
Required fields: image, prompt
Output fields: content_type (string, optional); url (string, required)
Execution mode: async

When to Use

Use the kling-video/v3/pro/motion-control variant of kwaivgi/kling-video/v3/pro/motion-control when the project specifically needs whether to keep the original sound from the reference video.