Kling 4.0
Kling 4.0 by Kuaishou creates 3–30 second videos at 720p or 1080p with synchronized stereo audio. Generate from text or images, guide motion with ordered keyframes and image, video or audio references, edit existing footage, or append a new segment with an optional ending frame.
Added Oct 21, 2026
Details
- Starting Price
- From $0.518 per video
- Model type
- both
Final price depends on the selected settings and is shown before generation.
Settings
Generation controls available for this model.
Output Format
Default Duration
5
29 duration options
Aspect ratio
Default
auto
Options (5)
auto, 16:9, 9:16, 1:1 +1 more
Generation mode
Default
image-to-video
Options (5)
Text to video, Image to video, Reference to video, Edit video +1 more
Middle frame URLs
Default
N/A
Image mode: ordered JPEG/PNG frames. Start, end and middle frames share a 10-image limit.
Original video sound
Default
auto
Options (3)
Default for mode, Keep original sound, Generate new sound
Output duration / new extension seconds
Default
5
Options (29)
Follow source (edit only), 3 seconds, 4 seconds, 5 seconds +25 more
Reference audio URLs
Default
N/A
Reference mode: up to 3 clips, 5–30 seconds each. Address clips as @Audio1, @Audio2. Images, videos and audios share a 15-file limit.
Reference image URLs
Default
N/A
Reference/edit mode: up to 10 JPEG/PNG images. One URL per line or a JSON array. Address references as @Image1, @Image2 in order.
Reference video URLs
Default
N/A
Reference mode: up to 5 videos; at most 30 seconds combined are used. Billing requires verifiable whole-second input duration.
Resolution
Default
1080p
Options (2)
720p, 1080p
Benchmarks
No public benchmark scores for this model yet.
Examples
Loading examples…