One motion clip, six colorways: Kling 3.0 Motion Control, $1.89 each
Kling 3.0 Motion Control bills $0.1575 per second. One 12-second motion clip applied to six product stills costs $11.34. The inputs and limits are inside.

Applying one 12-second motion video to six colorway stills of the same product costs $11.34 with Kling 3.0 Motion Control on Sume: $1.89 per clip at $0.1575 per second. The motion is identical across the six, which is the point when you sell one sweater in six colors.
The model is not text-to-video. A motion reference video you supply drives the movement of a still image, so the six outputs differ only in what the still shows.
Price arithmetic
The catalog records the Kling 3.0 Motion Control provider list at $0.126 per second, and Sume bills it at the standard 1.25 margin, which is $0.1575 per second. The charge is per whole second of motion video. Output length follows the driving video, from 1 to 30 seconds.
| Motion video length | One clip | Six colorways | Twelve colorways |
|---|---|---|---|
| 5 s | $0.79 | $4.74 | $9.48 |
| 8 s | $1.26 | $7.56 | $15.12 |
| 12 s | $1.89 | $11.34 | $22.68 |
| 30 s | $4.73 | $28.38 | $56.76 |
Request shape
Send exactly one visual source, either image_url or a ready avatar, plus motion_video_url and duration_seconds between 1 and 30. The declared duration is the reservation basis only, since the motion video sets the real length. keep_original_sound defaults to true, so set it to false for a silent clip if you will add your own bed. Reserve the same Idempotency-Key pattern per colorway so a retry never double-charges.
curl -X POST https://api.sume.com/v1/kling/3.0/motion-control \
-H "Authorization: Bearer $SUME_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: sweater-red-motion-01" \
-d '{
"image_url": "https://example.com/sweater-red.jpg",
"motion_video_url": "https://example.com/model-turn.mp4",
"duration_seconds": 12,
"keep_original_sound": false,
"character_orientation": "video"
}'
When this beats text-to-video
Text-to-video re-invents the motion on every call, so six colorways give six different performances and some will be worse. Motion Control fixes the performance and varies the look, so approval is one decision, not six. Your still is the visual source, which helps when a client will compare the clip with the product page.
Caveats
A person in the motion video is the person who moves, so the motion reference should be a clip you have the right to use. The prompt only steers appearance, and the route rejects Video 1.0 vocabulary such as reference_image_urls. Budget a retry per set: one re-render of the 12-second clip is $1.89.
Sources
Related posts
More in Use cases
- One narrator in many languages: a Sume TTS batch vs MAI-Voice-2.1
MAI-Voice-2.1 claims one voice across 23 languages. On Sume, you send the same voice.id with a language per request and audition each language.
- One Short in Korean and English: caption styles and rules
Posting a Korean and an English cut of your own AI Short? YouTube is silent on translations; Sume caption styles are not. Which style fits which script.
- One-take walkthrough prompts: store, kitchen, studio on Seedance 2.5
Three one-take walkthrough prompts for Seedance 2.5 (store aisle, restaurant kitchen, studio), each with a route, a pace and a 30 s price at 480p to 1080p.
- Open Enrollment Reminder Video With an AI Avatar: Script, Cost
Script an HR open-enrollment reminder as a Sume Avatar 1.0 clip: scenes, word and second counts, request JSON, and the cost on each quality tier.
Written by Sume