Kling 4.0 video edit rewrites a 3 to 30 second clip from a text instruction. It keeps the scene, the subject and the motion and changes what you describe, with up to 10 optional reference images.
Workbench
These examples were written for this endpoint. Replace the prompt and sample media URLs with your own accessible inputs, then check the current quote and model availability before submitting.
- Output5s
observatory-recast The source is our Kling 4.0 text-to-video sample of an astronomer with short silver hair turning the dome wheel and looking up. The prompt replaced her with the woman from @Image1 - copper-red wavy hair, a tan corduroy jacket over a black turtleneck - and asked to keep everything else. Frame by frame, her hand on the wheel, the turn of her head, the telescope, the dome and the crane up line up with the source; only the person changed, and the fingerless gloves went with the old outfit. The length was set to 5 seconds and the original sound was switched off, so the dome motor and the wind were generated new.
- 720p
- 5s
- Output6s
gas-station-snowfall The source is our Kling 4.0 image-to-video sample of a man with a road map leaning on a vintage convertible at a desert gas station at dusk. The prompt asked for heavy snowfall and to keep the man, his motion, the car and the camera. Snow lay on the hood, the canopy and the forecourt with flakes still falling, and his pose, the map, the car and the dolly move matched the source. The warm lamp turned into a cool white canopy light. With the length left on Match source, a 5.04 second clip came back at 6.04 seconds and the extra second continued his walk to the driver's door. The original sound was kept.
- 720p
This example starts at $0.113 per output second. The Playground updates the estimate when you change the settings.