Kling 3.0 on Sume: kling-3 is v3 Pro, motion control is separate
On Sume, kling-3 is Kling Video v3 Pro for prompt and frame video; Kling 3.0 Motion Control is a different surface billed on the driving video's seconds.
When you search for "Kling 3.0" on Sume you will meet two different things. kling-3 is listed as Kling Video v3 Pro and makes a clip from a prompt or from first and last frames, 4 to 15 seconds, at 720p or 1080p. Kling 3.0 Motion Control is a separate surface that copies motion from a motion reference video and bills on that video's seconds.
Pick by the input you have: a prompt or a still means kling-3; a clip with the move you want repeated means motion control.
Side by side
The two do not share a price card. The kling-3 row bills per output second with a sound-on and sound-off rate, while motion control bills one rate on the length of the driving clip.
| kling-3 (v3 Pro) | Motion Control (kling/3.0/motion-control) | |
|---|---|---|
| Input | Prompt, first frame, last frame | A driving video plus a still image |
| Length | 4 to 15 s | 1 to 30 s, follows the driving clip |
| Provider list rate | $0.112 per s silent, $0.168 per s with sound | $0.126 per s |
| Sume rate (list x 1.25) | $0.14 or $0.21 per s | $0.1575 per s |
| Reference inputs | None accepted | Takes a motion video as its one driver |
What kling-3 does not take
kling-3 accepts no reference_image_urls, reference_video_urls or audio references. A request that includes them fails with an unsupported-capability error. If you want a character to hold across shots, send the still as a first frame, or pick an id that takes references. Motion Control is not a workaround for that, because it needs a driving video.
Which one for a 12-second ad
A 12-second prompt-only clip on kling-3 costs $1.68 silent or $2.52 with sound at the Sume rate. The same 12 seconds of driving video through motion control costs $1.89. The two prices are close, so decide on the creative need: a new scene from a prompt, or a known move on a new subject. Neither one is a general stand-in for the other.
Quick chooser
If you are still unsure which Kling surface fits, use this.
Keep the two surfaces in separate code paths. They have different routes, different bodies and different price units, so sharing one request builder invites a field that one of them rejects.
- Prompt only, or one still as a first frame:
kling-3. - A clip whose motion you want to reuse on a new subject: Motion Control.
- Reference images to hold a character: neither; choose an id that takes references.
- A silent clip:
kling-3with sound off, at $0.14 per second.
Sources
Related posts
More in Models
- Longest single AI video clip by API in 2026: 30 seconds, two Sume ids
Seedance 2.5 and Wan 3.0 make a 30-second clip in one job on Sume. Kling 3 and MiniMax H3 stop at 15 s and Omni at 10 s. Limits and per-second prices.
- LTX-2.5 audio-to-video: three ways to drive a clip with audio on Sume
LTX-2.5 is not on Sume. To make video that follows a voice or track, use audio references on Seedance 2.5, Wan 3.0 or H3, or H3 Max lip sync. Limits and prices.
- Open-weights video wave: which ids Sume lists, and which it does not
Sume's v1 video catalog lists Seedance, Wan 3.0, MiniMax H3, Kling 3 and Omni 1.1. It has no LTX id and no weights or LoRA fields. How to check before you plan.
- Lyria 3.5 tempo and duration control: Flow Music versus the Sume API
Google says Lyria 3.5 gives easier tempo and duration control in Flow Music. On Sume the same model has no duration field, so you steer length in the prompt.
Written by Sume