Kuaishou's Kling AI has moved Kling 4.0 from preview to full availability on its site. The update focuses on control and length rather than just better-looking frames.
Creators can now generate a single native clip of up to 30 seconds and set as many as 10 keyframes to steer what a character is doing, when scenes change and how the plot moves. A single job can combine up to 10 images, 5 video clips and 7 subjects as references, so character, motion, camera and sound can all be guided together. Audio is generated with the picture rather than added afterwards, and multi-shot continuation helps keep characters consistent across scenes.
Kling has been showing the model to advertisers, including at Advertising Week in New York, which signals where it expects the money to come from: ads, short films and branded content that need a coherent sequence rather than a single impressive shot.
Our take
The AI video race is shifting from pretty to directable. Keyframes and consistent characters are what turn a demo tool into something a production team can schedule work around.
From the issue
China wants to export-control its own AI models, and that changes your stack Friday, October 16, 2026