Showcases

Kling 3.0 Omni Video Transformation: Source-Footage-Driven Shot Restructuring

A developer-facing demonstration of the transformation capability: input a 3–15.5 second source video, and Kling 3.0 Omni repaints the visual content while preserving the original clip's duration and aspect ratio, without generating an audio track. This entry uses source material of a "city street scene transformed into cyber neon" to show how the core input parameters are organized: video_urls accepts only a single MP4/MOV (≤200MB, 700–4553 px, 24–60 fps), resolution switches among 720p/1080p/4k, and the output aspect ratio and duration automatically follow the source clip; prompt describes the target visuals and camera-movement continuity, and can use @name to reference subjects in elements (2–4 images of the same person per subject, up to 7) to lock in the appearance of a person or product. Suitable for scenarios requiring "unchanged source structure, reworked visuals," such as secondary stylization of commercials, reskinning of e-commerce assets, and multi-version comparison of storyboard previews.

Style Video Transformation · restyle · short-form

Model Kling 3.0

Reference inputs

video_urls

Try it yourself

Prompt

Try this model

Transform the original street-shooting footage into a neon-soaked cyberpunk night scene: rain-wet asphalt reflects magenta and cyan sign lights, dense holographic billboards, steam drifting from vents. Keep the original camera path, subject positioning, and rhythm exactly consistent with the source clip; re-render only materials, lighting, and atmosphere. Maintain the identity of the walking subject referred to as @walker throughout the shot, preserving natural gait and cloth movement, and keep the lens characteristics of soft anamorphic widescreen flares, shallow depth of field, 24 fps motion cadence, and cinematic color grading with deep blacks and highly saturated highlights.

Copy the full prompt and adapt it for your own generation.