Developed by Kuaishou, Kling AI 2.6 Motion Control is a performance-driven image-to-video model. It precisely transfers real movements, poses and expressions from a reference video to a character image, producing temporally stable, realistic video.
Open Motion WorkspaceDeveloped by Kuaishou, Kling 2.6 Motion Control transfers a real person's movement, gestures and facial expressions from a reference video to a character in a still image. It focuses on reusing performance rhythm, emotional expression and body language rather than a manual animation production process.
The Standard mode focuses on efficient video generation, with motion transfer and character performance for marketing materials, explainers, educational content and social media shorts. Standard output is 720p.
Open workspacePro prioritizes higher visual quality and finer rendering while following the same motion patterns. It supports cinematic scenes, virtual presenters, narrative content and brand videos requiring greater image detail. Pro output is 1080p.
Open workspaceKling AI 2.6 Motion Control
Kling 2.6 Motion Control transfers full-body movement from a reference video to a character image, including posture, movement rhythm and limb coordination. This supports performances involving large movements and dynamic scenes.
Kling AI 2.6 Motion Control reproduces coordinated movement involving multiple limbs, with a focus on structure and fluidity. This supports detailed character movement with continuity across the sequence.
Kling AI Motion Control captures and reproduces hand and fingertip movements from a reference video, including pointing, grasping objects and expressive gestures for product demonstrations and dialogue videos.
Kling Motion Control supports continuous movement up to 30 seconds for narrative scenes, demonstrations and sustained action sequences. Reference-video orientation supports 3–30 seconds; character-image orientation is limited to 10 seconds.
The reference provides movement and expressions, while the prompt defines scene details. Kling AI 2.6 Motion Control allows the background, environment and contextual elements to be defined separately, so one performance can be reused in different visual settings.
The Kling AI Motion Control allows teams to reuse a single human performance across multiple brand characters or spokespersons. By transferring the same motion and expression to different visual identities, marketers can produce consistent, on-brand videos for campaigns, product launches, and social media without repeated filming or manual animation.
With the Kling 2.6 Motion Control, presenters’ gestures, hand movements, and pacing are preserved while character appearance and background can be customized. This makes it well suited for product demos, app walkthroughs, and explainer videos where clear gestures and natural presentation flow are essential.
The Kling Motion Control enables realistic motion-driven content for AI influencers and virtual creators. Real human performances can be mapped onto virtual personas, maintaining natural body language and expression while allowing creators to scale content production across platforms such as short-form video, livestream clips, and UGC-style media.
For training and educational content, the Kling AI 2.6 Motion Control helps deliver consistent instruction by reusing instructor performances across different characters or scenes. Gestures, posture, and expression remain stable, making it suitable for onboarding videos, internal communications, and online learning materials that require clarity and engagement.
Provide one character image and one motion video. The model uses the image for the subject and the video for movement; it cannot start from a text prompt alone. Keep the subject’s head, shoulders and torso clearly visible.
Use JPEG or PNG up to 10 MB, with dimensions greater than 300 px and aspect ratio from 2:5 to 5:2. Avoid using a completed video in this image input; it can only be previewed there.
Use MP4 or MOV, 3–30 seconds and up to 100 MB. Image-based orientation limits generation to at most 10 seconds; video-based orientation allows up to 30 seconds. Choose the orientation for the length and framing you need.
Image keeps the orientation aligned to the character picture. Video follows the character orientation in the performance clip. This also affects the maximum duration, so review it before submitting a long motion reference.
Both qualities are available. The character image and motion clip have the same reference requirements for either setting. Compare the quote after choosing a quality and motion reference.
The current 2.6 workflow exposes orientation and quality but does not provide a separate background-source control. Kling 3.0 Motion Control offers that additional choice.
A prompt can describe the intended scene in up to 2,500 characters. The reference video supplies the performance, so use text to clarify the result rather than replacing the required image or video.
Developed by Kuaishou, Kling AI 2.6 Motion Control is a performance-driven image-to-video model. It precisely transfers real movements, poses and expressions from a reference video to a character image, producing temporally stable, realistic video.
Open Motion Workspace