Kling 4.0 supports up to 15 combined reference assets, including images, videos, subjects, and voice references, subject to individual input limits. Combine character images, scene references, and motion footage to guide appearance, camera work, movement, and visual style within the same generation.
You can use existing footage pacing, composition, and movement as references for new recreation. Kling 4.0 also supports targeted video editing, including changes to characters, products, backgrounds, camera angles, and visual style. Improved consistency helps preserve recognizable subjects across different scenes and creative variations.