KLING 4.0
Kling 4.0 — direct the scene, references and sound
Shape 3–30-second videos from text, images or video references, with native audio and a separate video-editing mode.
Watch original Kling 4.0 scenes with their prompts, then direct your own video in App, Studio or MCP. Each generation uses a current quote and requires your confirmation.
Direct a sequence
Use one prompt for a continuous scene, or describe timed shots whose durations add up to the full clip. Give each shot one clear action.
Guide the subject and setting
Start from an image, add an optional ending image, or use reference images, videos and elements to guide a new scene.
Write sound into the scene
The provider announces native dialogue, singing and stereo audio. Describe the words, speakers and atmosphere in the prompt; no separate audio switch is documented.
Edit an existing video
The editing mode takes a source video and a prompt, with optional reference images and image elements. It has a different input contract from reference generation.