Kling's newest video model, puts realism, control and full stories first. It generates up to 30 seconds in a single take with stereo sound and lip-synced dialogue, follows up to 10 key frames or 15 references in one prompt, and edits footage you already have, all at 720p or 1080p on Pika.
What you can create
Long takes, one pass
Anywhere from 3 to 30 seconds as one continuous clip. Line up as many as 10 key frames to fix the beats, and Kling fills in the motion between them.
Direct with references
Combine up to 10 images and 5 clips, 30 seconds of video in total, and @mention each one. Borrow a proven video's pacing and framing, then swap in your own product or character.
Camera moves that hold steady
Push-ins, pans, tracking shots and orbits stay smooth, and busy, fast-moving scenes stay stable from start to finish.
Sound and speech built in
Every clip comes with two-channel stereo and tighter lip sync. Characters can speak Chinese, English, Japanese, Korean, Spanish, Portuguese, German, French or Hindi, with regional accents.
Readable text on screen
Signs, captions, logos and emoji come out legible in many languages and stay clear as the camera moves.
Edit footage you already have
Open a clip in Edit Video and choose Kling 4.0 to change an expression, a movement, the camera, the style or the background while the rest of the scene stays put.
Examples
A desert-road long take
Roller skates, tracking shot
Directed from references
From a still: wind over the field
From a still: a snake on the bed
From a still: the long dinner table
Frequently Asked Questions
How long can a Kling 4.0 video be?
Any whole number of seconds from 3 to 30, generated in one take.
Which resolutions and aspect ratios can I use?
720p or 1080p on Pika, in 16:9, 9:16, 1:1 or 21:9. In Key Frames the clip takes the shape of your frames.
Does Kling 4.0 make audio?
Yes. Every clip comes with two-channel stereo audio, including lip-synced dialogue, effects and music, so there is no audio switch.
What is the difference between Omni Reference and Key Frames?
Omni Reference treats what you attach as ingredients you describe in the prompt. Key Frames treats each image as a moment in the story, played in the order you set.
What can I edit with Kling 4.0?
Expressions, body movement, camera position and movement, overall style and backgrounds. Each clip can be 3 to 30 seconds, and the clips in one edit add up to 30 seconds at most.
Can I still use Kling 3.0?
Yes. Switch the Model row to Kling 3.0 for up to 4K output and an optional audio switch.
