Kling motion control vs Genjutsu Motion Transfer: what each costs
Both repeat a reference video's movement. Kling motion control: one photo, $0.1575 a second. Genjutsu: 1 to 8 images, $0.3975 (480p) or $0.85125 (720p).
The short answer
Both routes repeat the movement of a reference video, but they bill and behave differently. Kling motion control takes one character photo and costs $0.1575 per second. Genjutsu Motion Transfer takes one source video plus 1 to 8 images and costs $0.3975 per second at 480p or $0.85125 at 720p. A 10-second clip: $1.575 versus $3.975 or $8.5125.
Same 10 seconds, two routes
Rates are billed per second at list x 1.25. A 10-second reference clip prices as follows.
| Route | Setting | Per second | 10 seconds |
|---|---|---|---|
| Kling motion control | - | $0.1575 | $1.575 |
| Genjutsu Motion Transfer | 480p | $0.3975 | $3.975 |
| Genjutsu Motion Transfer | 720p | $0.85125 | $8.5125 |
How the inputs differ
Kling motion control: one image_url (or an avatar), one motion_video_url, 1 to 30 seconds, and keep_original_sound for the audio. The prompt describes appearance only. It is a dedicated endpoint, POST /v1/kling/3.0/motion-control.
Genjutsu
Genjutsu goes through the generic video routes: /v1/videos with input_references containing exactly one video_url and 1 to 8 image_url entries, or /v1/video-router/generate with video_url and reference_image_urls. The duration must be the input video's inspected length rounded up, 4 to 30 seconds. The output keeps the source length and framing, and there is no audio output. It is a transfer, not a new shot.
Which one
Genjutsu is available only where its provider is configured, so a catalog check comes first. Kling motion control is the cheaper default for one character and one move. Reach for Genjutsu when you have several reference images and need the output to match the source video's length and framing exactly.
Test small
Whichever you choose, test with a short slice of the reference first. Both bill by the second, so a 4-second test is a fraction of the final.
Sources
Related posts
More in Comparisons
- Ray3.2's 16 keyframes vs Omni's two frames: what Sume can do
Luma lists up to 16 keyframes for Ray3.2; Omni takes two. On Sume you chain Omni pairs: 5 keyframes is 4 clips of 5 s, $3.76 at 1080p.
- Ray3.2 takes 16 keyframes per clip. Sume takes two frames: chain them
Luma lists up to 16 keyframes per Ray3.2 clip. Sume does not ship Ray3.2, but its first and last frame jobs chain into a keyframed shot.
- Luma Ray3.2 20 s 1080p vs MiniMax H3 Max 15 s 1080p on Sume
Luma lists Ray3.2 at 1080p for up to 20 s with 16 keyframes. Sume does not run Ray3.2; H3 Max 1080p is 5-15 s at $0.20 a second. What to use for 20 seconds.
- Luma Ray3.2 features against what Sume's docs list: a one-page matrix
Luma lists 16 keyframes, 20 s, 1080p, HDR with 16-bit EXR, 8-face tracking and Reframe for Ray3.2. Sume does not ship it; here is what Sume's docs cover.
Written by Sume