AI Models
What each model is good at, where it differs from the others, and what it costs to run.
Kuaishou's Kling 3.0 motion transfer. One picture of a subject plus one clip whose motion it copies. A prompt is optional.
Credits come with a Generator plan. The exact cost is shown on the Generate button before anything is spent.
Kling 3.0 Motion Control is Kuaishou's motion-transfer model on the Kling 3.0 line. The picture is the subject. The clip is the performance.
Kuaishou describes Kling 3.0 motion control the same way as on 2.6: full-body motion from a reference clip, including lip-sync and expression, applied to a character image. The 3.0 line is the one Kuaishou launched for stronger consistency and native audio in ordinary clips. This model is the motion-transfer member of that line.
Attach one picture of the subject and one clip. A prompt is optional.
Kling 2.6 Motion Control takes the same inputs on the previous Kling line. Kling 3.0 generates from a prompt or from opening and closing frames. It does not copy a performance. Genjutsu Motion Transfer carries a clip's motion onto one to eight pictured subjects.
One picture and one clip. The prompt is optional.
No. You need the picture and the clip.
Same inputs. This is the Kling 3.0 model. Kling 2.6 Motion Control is the earlier one.
No. Omni Reference builds a clip from several pictures. Motion control copies a clip's movement onto one pictured subject.
