Lip sync and stereo
The singer's mouth follows the vocal while the studio recording plays.
Kling 4.0 AI Video
Kling 4.0 is Kuaishou's next video model, planned for October 2026. One generation can run 3 to 30 seconds, with up to 10 keyframes, 15 references, and output from 720p to 4K. MixVio does not generate Kling 4.0 yet.
- Kuaishou
- Text, image, or references
- Native 30-second video
- 720p to 4K
Kling 4.0 Video Examples
Twelve official films from the September 28, 2026 release note. These show the look. Key Features below use different clips.
Ultrawide lobby action
Prompt
Ultrawide hotel lobby video where two people struggle beside a purple chair.
Key Features of Kling 4.0
Omni Reference
The red supercar and the motorcycle share one shot, so two subjects stay in the same scene.
Story beats
The shot leaves a glowing star in an open hand and picks up a creature in the air.
What is Kling 4.0?
Kling 4.0 is the next Kling video model from Kuaishou. Kling announced it on September 28, 2026, and plans the full model for October 2026.
What Kling announced
- Native video from 3 to 30 seconds in one generation
- Up to 10 keyframe images
- Up to 15 reference items through Omni Reference
- 720p, 1080p, and 4K, with 10-bit HDR coming soon
- Two-channel stereo and tighter lip sync
- Video extension up to 2 minutes, marked coming soon
When can I use Kling 4.0?
Kling says the full model is planned for October 2026. Kling 4.0 Flash opened to a limited group on Kling on September 28, 2026, at 3–20 seconds, 720p, and 8-bit SDR. MixVio does not generate Kling 4.0 yet.
Create with Kling 3.0Kling 4.0 at a glance
Picture and sound
Steadier large motion, two-channel stereo, tighter lip sync, and output up to 4K. 10-bit HDR is still marked coming soon.
Omni Reference
Up to 15 items across images, short videos, a voice reference, and subjects, with the same inputs available for video editing.
Longer stories
Native 3–30 second generation, guided by up to 10 keyframes.
Languages and text
The spec table lists major languages, accents, and dialects, plus on-video text and emoji.
Who is Kling 4.0 for?
Longer native video, keyframes, and references matter most when a story needs more than one shot.
Filmmakers and storyboard artists
Place up to 10 keyframes and let one 3–30 second generation carry a scene.
Product and commerce teams
Hold a product, person, and setting with image, video, and voice references.
Music and performance videos
Plan performance shots around two-channel stereo and tighter lip sync.
Ad and brand studios
Direct a spot with references and keyframes, then edit a shot instead of starting over.
Short drama makers
Use native 30-second scenes and repeated references when a story continues across shots.
Multilingual creators
The published language list includes accents and dialects for versions aimed at different markets.
Kling 3.0 Pricing
Compare Kling 4.0How Kling 4.0 compares to Kling 3.0
Kling 4.0
Announced- Single generation
- 3–30 seconds
- Story control
- Up to 10 keyframes
- References
- Up to 15 items
- Resolution
- 720p, 1080p, and 4K
- Audio
- Two-channel stereo and tighter lip sync
Kling 3.0On MixVio
- Single generation
- Up to 15 seconds
- Story control
- Multi-shot storyboards
- References
- Image and video references
- Resolution
- Up to 4K
- Audio
- Native audio
Source: Kling 4.0 release note
Kling 4.0 specs
Figures follow the September 28, 2026 release note. HDR and the two-minute extension stay marked as coming soon.
- October 2026
Full release
Kling says the full model is planned for October 2026.
- 3–30 seconds
Native length
One generation. Extension up to 2 minutes is coming soon.
- 720p · 1080p · 4K
Resolution
10-bit HDR is coming soon on 1080p and 4K. Ratios: 21:9, 16:9, 1:1, 9:16, Auto.
- Up to 10 images
Keyframes
Also listed: text to video, image to video, and first and last frames.
- Up to 15 items
References
Up to 10 images, 5 videos totaling 30 seconds, voice only, and up to 7 subjects, including up to 3 video subjects.
- Stereo
Audio
Two-channel stereo, with tighter lip sync.
- 8,000 tokens
Prompt length
Published prompt limit for Kling 4.0.
- And more
Languages
Chinese, English, Japanese, Korean, Spanish, Portuguese, German, French, Indonesian, and more, including accents and dialects.
- Flash
Kling 4.0 Flash
Limited access on Kling since September 28, 2026. Published limits are 3–20 seconds, 720p, and 8-bit SDR.
- Coming soon
10-bit HDR
Coming soon on 1080p and 4K.
- Coming soon
Video extension
Coming soon, up to 2 minutes.
- Not on MixVio
Generation
MixVio will add Kling 4.0 after the official API, price, and request fields exist. Kling 3.0 is available now.
Kling 4.0 FAQs
What is Kling 4.0?
Kling 4.0 is the next Kling video model from Kuaishou. Kling announced it on September 28, 2026. Published limits include native video from 3 to 30 seconds, up to 10 keyframes, up to 15 references, and output at 720p, 1080p, and 4K. The full model is planned for October 2026.
When can I generate Kling 4.0 on MixVio?
Not yet. MixVio will add Kling 4.0 after the official API, price, and request fields are available. Kling 3.0 text-to-video and image-to-video are available on MixVio now.
Does MixVio list a Kling 4.0 price?
No. Kling 4.0 has no MixVio price yet. The plan cards on this page are the live Kling 3.0 quote.
What inputs does Kling 4.0 accept?
The release note lists text to video, image to video, first and last frames, up to 10 keyframe images, and Omni Reference. Omni Reference can combine images, videos, and a voice reference within the published caps. Video editing accepts up to 5 source videos, with one primary video.
How long can a Kling 4.0 video be?
Kling says one generation can run from 3 to 30 seconds. Repeated extension up to 2 minutes is marked coming soon. Kling 3.0 generations stop at 15 seconds.
Start your Kling 3.0 scene.
Private. The quote is shown first.
Start from a text prompt or an image, then generate with Kling 3.0 selected.
Create with Kling 3.0