Make a still dance to a 22-second reference: $3.465 on Sume
Kling 3.0 Motion Control copies the movement of a 22-second dance video onto one character still. On Sume it costs 22 x $0.1575 = $3.465.

To make a character still dance like a 22-second reference video, call Kling 3.0 Motion Control on Sume with the still and the dance clip. It costs 22 x $0.1575 = $3.465, and the output is as long as the driving video.
What you send
You need two public HTTPS URLs: image_url for the character and motion_video_url for the dance. You can send avatar_handle instead of image_url if you already created an avatar; you cannot send both. You also declare duration_seconds, here 22, so that the reservation matches the clip.
- Still: one image of the character, ideally full body if the dance uses legs.
- Driver: the 22-second dance, from 1 to 30 seconds long.
- Optional prompt: up to 2,000 characters.
- Price: ceil(22) = 22 seconds x $0.1575.
Cost against other dance lengths
The price grows by $0.1575 for every second of dance. The table shows four common lengths.
| Driver | Arithmetic | Cost |
|---|---|---|
| 9 s | 9 x $0.1575 | $1.4175 |
| 15 s | 15 x $0.1575 | $2.3625 |
| 22 s | 22 x $0.1575 | $3.465 |
| 30 s | 30 x $0.1575 | $4.725 |
Planning notes
A 45-second routine does not fit one job, because the limit is 30 seconds. Split it into two drivers (for example 22 s and 23 s, which cost $3.465 and $3.6225) and join the results with the timeline route. Sume bills the render by output minute, so the join adds a small amount on top.
Before you publish, remember that you are copying a performance. If the dance is someone else's choreography or a video you did not film, get permission to use it as a driver, and if the character looks like a real person, follow the platform's disclosure rules for synthetic content.
Preparing the still and the driver
The still decides who dances and the driver decides how. Use a full-body still when the dance uses legs, with arms visible and the background simple, since the output follows the framing of the driving video. The docs give no minimum image size, so I will not state one; test with your usual asset on a 3-second driver at 3 x $0.1575 = $0.4725.
Host both files at public HTTPS URLs. The route refuses localhost, private network and non-HTTPS links. If you create the character with Sume first, pass its avatar_handle in place of the image URL.
If the dance includes music, decide up front whether to keep the original sound. The default keeps it, and the soundtrack's rights remain your responsibility. For a silent result that you score yourself, set keep_original_sound to false.
Sources
Related posts
More in Use cases
- 24 children's book pages at 4K: $4.80 or $5.34 on Sume
Twenty-four 4K pages cost $4.80 on Nano Banana 2.1 or $5.34 on GPT Image 2.5 high on Sume. Google lists Nano Banana 2.1 4K at $0.113 direct.
- 250 product photos to 5-second 1080p clips with Omni: $234.38 on Sume
250 SKUs as 5-second image-to-video clips cost $234.38 at 1080p, $156.25 at 720p and $46.88 at 360p on Sume's Omni row. A draft plan brings it to $103.13.
- 250 SKUs, two 4K heroes each: $100.00 or $111.25 on Sume
Five hundred 4K retail images cost $100.00 on Nano Banana 2.1 or $111.25 on GPT Image 2.5 high on Sume. At 2K, $75.00 or $27.81. Arithmetic inside.
- 28-minute podcast episode on Sume: narration, bed, render = $4.125
A 28-minute narrated episode costs about $4.125 on Sume: two TTS requests ($1.20), one Music 1.0 bed ($0.125) and a 28-minute Timeline render ($2.80).
Written by Sume