Kling 4.0 is the flagship video model from Kling AI. It creates realistic motion and synchronised sound together, and a single generation can run anywhere from 3 to 30 seconds. That length gives a scene room to develop: a slow reveal, a short story with several beats or a full product demo, without stitching clips together.
On AdMultiplier, Kling 4.0 works in three modes. Text mode turns a written scene into video. Frame mode starts from your image and can end on a second one. Reference mode accepts up to 10 images, 5 videos and 3 voices, 15 items in all, which you call by name in the prompt as @image_1, @video_1 or @voice_1, so a product, a presenter and a voice all land in the same clip.
Every Kling 4.0 video comes with native stereo sound, including voices that match the lip movements. Choose 720p or 1080p and 16:9, 9:16, 1:1 or 21:9. Without reference videos the rate is 57 credits per second at 720p and 85 at 1080p. With reference videos it is 34 or 51 credits per second, counted on twice the output length or on output plus reference seconds, whichever is longer.