Mixer AI
Kling 2.6 Motion Control
A photo supplies the character; a video supplies the performance, from dance to conversational gestures.
When to choose it
Kling 2.6 Motion Control is for movement you have already recorded and would rather show than describe. Upload a character image and a performance clip. Match the framing: a full-body dance needs a full-body character, while conversational gestures work with waist-up framing.
- Use an existing performance as the movement guide.
- Choose whether the photo's pose or the video's angles should guide character orientation.
Modes and settings
- One photo and one video as inputs; 720p or 1080p output.
- Video orientation supports up to 30 seconds; image orientation up to 10 seconds.
Price in coins
| Quality / size | Input | Coins / second |
|---|---|---|
| 720p | Photo + motion video | ≈ 7.62 |
| 1080p | Photo + motion video | ≈ 12.21 |
Billing is per second, not per fixed-length clip. The rate is multiplied by the billable duration; the complete request total is rounded up to a whole coin once.
Billing uses the duration of the motion video transferred to the character.
≈ marks a fractional rate rounded to two decimal places for display; billing uses full precision.
What you can make
- Transfer a short dance to an illustrated character.
- Give a portrait gestures from your own recording.
How to direct the result
- Assign appearance to the photo and motion to the video instead of rewriting the entire choreography.
- Leave room for arms and legs. Start with a short, clearly visible performance.
Example: a greeting gesture
Take appearance and clothing from the photo and movement from the video. The character raises their right hand, gives a brief greeting and lowers it. Preserve the face and proportions. No additional people or actions; keep the surroundings calm.
What to keep in mind
- A written prompt cannot replace the required motion video.
- Occluded limbs, sharp turns and movement outside the frame can transfer incorrectly.
Sources
- Prices and availability — Mixer AI catalog, updated 2026-09-18.
- Official maker site: Kuaishou
Model facts verified: 2026-09-20.
Guides and comparisons using this model
FAQ
Is one photo enough?
No. The motion clip is also required because it provides the performance.
What makes a useful dance reference?
One clearly visible performer, no cuts and no cropped hands or feet.
Why choose character orientation?
It determines whether the source photo or video has more influence over pose and viewing angles.
Can I make twenty seconds?
Video orientation allows up to 30 seconds. Image orientation stops at 10 seconds.
Can I use an illustration?
Try a character image with clear anatomy and pose. A large difference from the performer makes transfer harder.
Why does the face change?
Check angle and framing compatibility. Start with smaller head turns and footage without facial occlusions.
What changes the price?
The coin rate per second depends on quality, mode and, for some models, sound. Multiply it by the billable duration. Where input video is charged, its duration also counts. The complete request total is rounded once.
Do I need a separate model subscription?
No. In Mixer AI, you top up one coin balance and use it for the tasks you need. You do not need a separate subscription to this model.