Kling 4.0: 30-Second AI Video with Sound

Kuaishou's fourth-generation Kling, now on VidCella. Up to 30 seconds per clip with native stereo audio, first, middle and last frame control, and up to 10 image and 5 video references. 720p or 1080p, from 43 credits/s.

Key Capabilities

What Kling 4.0 Adds

Longer clips, sound out of the box, and references that carry a person or a camera move into a new shot.

Put a Real Person into a New Shot

Upload a photo of someone and a clip with the camera move you want. Here the woman from a café portrait ends up rowing across a misty lake, following the reference video's motion.

A Sequence of Actions in One Take

Write the beats in order and they play out in order: she laughs, rests her chin on her hand, then grabs her bag and leaves. The café ambience is generated with the picture.

Atmosphere You Can Hear

Fog, water and birds move the way they should, and the sound comes from the same prompt: lapping water, distant calls, no separate audio pass.

What Is Kling 4.0?

Kling 4.0 is the fourth major generation of Kuaishou's Kling video model, released in October 2026. It renders single clips up to 30 seconds with stereo audio, from text, images, keyframes or reference videos. On VidCella both of its modes live on one page.

Text / Image to Video
Prompt only for text to video. Add a first frame, up to 8 middle keyframes and a last frame to steer where the shot goes.
Reference to Video
Mix up to 10 reference images and 5 reference videos, and point to each in the prompt as @image_1 or @video_1.
3 to 30 Seconds
Any whole-second length in that range, long enough for a full scene rather than a single beat.

How to Use Kling 4.0 on VidCella

Idea to finished clip in four steps:

Kling 4.0 Features on VidCella

What you get, pay-per-generation:

3-30 Second Clips

Any whole-second length, at 720p or 1080p.

Native Stereo Audio

Dialogue, ambience and effects are generated together with the video.

Up to 10 Frames

First, middle and last frames. Longer clips allow more of them, up to 10 at 9 seconds and above.

Image & Video References

Up to 10 images and 5 videos (30 seconds combined), with the option to keep the reference videos' own sound.

Four Aspect Ratios

16:9, 9:16, 1:1 or 21:9 ultrawide. Reference to Video can match your references automatically.

Pay by the Second

43 credits/s at 720p, 64 at 1080p. No subscription, no queue, no waitlist.

FAQ

FAQs about Kling 4.0

Common questions about Kling 4.0 on VidCella

What is Kling 4.0?

Kuaishou's fourth-generation AI video model, released in October 2026. It generates clips up to 30 seconds long with native stereo audio, from text, images, keyframes and reference videos.

How much does Kling 4.0 cost on VidCella?

43 credits per second at 720p and 64 at 1080p, so a 5-second 720p clip is 215 credits. With reference videos the rate is 26 (720p) or 39 (1080p) credits per billed second, and billed seconds are the larger of twice the output length or input plus output length: a 5-second 720p clip with a 3-second reference video costs 260 credits. The exact cost is shown before you generate.

How long can a clip be?

Anywhere from 3 to 30 seconds in a single generation.

Does it generate sound?

Yes, every clip comes with audio. In Reference to Video you can keep the original sound of your reference videos instead.

How do I keep the same character between shots?

Use Reference to Video. Upload a clear photo of the person, mention it as @image_1, and describe the new scene. Add a reference video if you also want its camera movement.

Kling 4.0 or Kling 2.5 Turbo?

Kling 4.0 runs up to 30 seconds with stereo audio, middle keyframes and video references. Kling 2.5 Turbo is the cheaper pick for silent 5 or 10 second drafts, at 60 or 120 credits a clip.

Can I use the videos commercially?

Yes. Handle real-person likeness, trademarks and third-party IP responsibly under applicable law and our terms.