A fierce ceremonial sword-dance performance in a burning neon temple, elegant and precise.
Kling 3.0 Motion Control API
Kling 3.0 Motion Control transfers body movement and action from a reference video to the character in your uploaded image.
Upload one character image and one motion reference video up to 30 seconds. Prompt guidance is optional.
Model Type:Motion control
Pricing
Motion control
The Standard rate is multiplied by the input video duration, rounded up to the next whole second.
- Standard
- $0.09/s
Playground Examples
Try production-style prompt starters.
An otherworldly ritual dance with fluid, hypnotic motion as the crystal cavern pulses with light.
A sharp futuristic runway dance with explosive turns and controlled robotic precision.
Similar Models
Compare adjacent capabilities before switching models.
Model overview
Kling 3.0 Motion Control for reference-driven character animation
Use a character image and a motion reference video to create a new clip whose subject follows the reference movement while retaining the uploaded appearance.
Character image plus motion reference
The image defines the character appearance while the reference video supplies body movement, pose, and action timing.
- Supports Motion control and reference-asset generation workflows.
- Fits product assets, ad previews, visual direction, and social content tests.
- Adjust configured inputs, media, and switches before generation.
Reference clips up to 30 seconds
The current video-orientation route accepts reference videos up to 30 seconds and measures their duration before task creation.
Character imageMotion reference video
Focused public controls
The public surface exposes the required image and video plus an optional prompt while execution routing remains fixed internally.
- Compare available models from Google, OpenAI, ByteDance, Kuaishou in one place.
- Copy the request body directly to the server to reduce frontend/backend parameter drift.
- Failure states, retries, result preview, and review checkpoints stay in the same workflow.
Model characteristics
Built for motion transfer from video to video.
Motion transfer
Apply action and pose timing from the reference video to a new character.
Appearance guidance
Use one image to define the generated character or subject.
Optional prompt
Add scene or motion guidance only when the reference assets need clarification.
FAQ
These are the first questions to answer when evaluating this model.
How long can the reference video be?
The current video-orientation route accepts reference videos up to 30 seconds.
Can I select Standard or Pro?
No. The current sjolt route uses the fixed Standard execution path.
Can I control character orientation or original audio?
No. Those controls are not exposed, and audio handling is provider-defined.