Kling 4.0 AI Video Generator
Kling 4.0 is the next Kling video model: single shots of up to 30 seconds, output up to 4K, stereo audio with more accurate lip sync, and up to 15 references in one request.
Kling 4.0 is not on Veevid yet. We'll add it once its API opens. Until then, the generator below runs Kling 3.0.
Text to Video
What's New in Kling 4.0
The headline features of the full Kling 4.0 model, coming to Veevid once its API opens.
30 second single shots
One continuous shot runs anywhere from 3 to 30 seconds, twice the 15 second ceiling of Kling 3.0. Extending a video several times, up to 2 minutes, is listed as coming soon.
Up to 4K output
Render at 720p, 1080p or 4K in 21:9, 16:9, 1:1, 9:16 or Auto. 10-bit HDR at 1080p and 4K is listed as coming soon.
Stereo audio and multilingual lip sync
Sound comes back in two channel stereo with more accurate lip sync, in languages including Chinese, English, Japanese, Korean, Spanish, Portuguese, German, French and Hindi, plus a range of accents and dialects.
Omni Reference with up to 15 inputs
Mix up to 10 images, 5 video clips (30 seconds in total) and 7 elements, up to 3 of them built from video. Audio can be added as a voice reference.
Up to 10 keyframes
Set a first and last frame, or pin up to 10 keyframes along the shot, and the model fills in the motion between them. Prompts can run up to 8,000 tokens.
Video editing
Edit with up to 5 input clips: one main video, each clip 3 to 30 seconds and 30 seconds in total. Change expressions, actions, camera moves, style or background, and add or remove elements.
Kling 4.0 vs Kling 4.0 Flash vs Kling 3.0
The specs side by side.
| Spec | Kling 4.0 | Kling 4.0 Flash | Kling 3.0 |
|---|---|---|---|
| Length per generation | 3 to 30 seconds per shot | 3 to 20 seconds | 3 to 15 seconds |
| Resolution | 720p, 1080p, 4K (10-bit HDR at 1080p and 4K coming soon) | 720p, 8-bit SDR | 720p, 1080p, native 4K |
| Aspect ratios | 21:9, 16:9, 1:1, 9:16, Auto | Not listed | 16:9, 9:16, 1:1 |
| References | Omni Reference, up to 15: 10 images, 5 clips (30 s total), 7 elements, voice audio | Omni Reference | Omni: up to 7 images and 1 clip of 3 to 10 s |
| Frame control | First and last frame, or up to 10 keyframes | First frame only | First and last frame; multi-shot up to 6 shots |
| Audio | Stereo, more accurate lip sync, 9+ languages plus accents and dialects | Not listed | Native audio in 5 languages |
Kling 4.0 and Kling 4.0 Flash specs are from Kling AI's release note. The Kling 3.0 column shows Kling AI's official specs; on Veevid, Kling 3.0 runs text to video and image to video at 3 to 15 seconds in Standard or Pro, without 4K output or Omni Reference.
How to Get Ready for Kling 4.0
Kling 4.0 is not on Veevid yet. Here is how to use the time until it arrives.
Draft with Kling 3.0 today
The generator on this page runs Kling 3.0. Use it to test prompts, framing and source images on clips of up to 15 seconds.
Plan longer shots and references
Kling 4.0 takes up to 30 seconds, 10 keyframes and 15 references. Map out the beats of a longer shot and collect the images and clips you want to reference.
Come back when Kling 4.0 lands
We plan to add Kling 4.0 to the model picker on this page once its API opens. Bring your tested prompts and assets over and render the longer version.
Kling 4.0 FAQ
Common questions about Kling 4.0, Kling 4.0 Flash and Kling 3.0.
Use Kling on Veevid Today
Kling 3.0 and Kling Motion Control are available now.
Start with Kling 3.0 while Kling 4.0 is on the way
Kling 3.0 is live on Veevid now. Draft your shots today and come back here when Kling 4.0 arrives.
Start With Kling 3.0