Generate video with Kling 4.0 from a prompt, a first frame, first and last frames, or reference images and videos. Every clip comes with its own sound, up to 30 seconds long. Start from the sample prompts below.
Both clips below were made with Kling 4.0 at 720p, 3 seconds. The previews play muted. Tap Recreate to load the exact prompt into the generator above.
16:9 · 3sA barista in a small sunlit cafe slides a cup of latte across the counter, smiles at the camera and says: "Your coffee is ready." Gentle cafe ambience, cups clinking.
Kling 4.0 generates native audio with every clip: ambience, effects and spoken lines. Put dialogue in quotes in the prompt and the character says it, with matching lip movement.
9:16 · 3sVertical selfie-style shot: a young woman hiking on a windy mountain ridge at golden hour turns to the camera, grins and says: "We made it to the top!" Her hair and jacket flutter in the wind, clouds drift through the valley below, wind rushing.
Render 3 to 30 seconds in 16:9, 9:16, 1:1 or 21:9 at 720p or 1080p. With a first frame, the clip keeps that image's shape.
| Resolution | 720p and 1080p |
|---|---|
| Length per clip | 3 to 30 seconds per clip |
| Frame shapes | 16:9, 9:16, 1:1 and 21:9; frame tasks follow the first frame |
| Audio | Native audio, including speech; always on without a reference video |
| What it takes | Text, a first frame, first and last frames, or up to 10 reference images and 5 reference videos |
Both are available here. What changes for you:
| At a glance | Kling 4.0 | Kling 3.0 |
|---|---|---|
| Video length | 3 to 30 seconds | 3 to 15 seconds |
| Sound | Always generated, at no extra cost | Optional Sound switch, costs more |
| References | Up to 10 images and 5 videos | First and last frame only |
| Highest resolution | 1080p | 4K |
Describe the subject, the action, the camera and the light, then the sound. Write any spoken line in quotes.
Leave the frame empty for text to video, add a first frame (and a last frame) to control the shot, or switch to Reference to bring in up to 10 images and 5 videos and mention them as @Image1 or @Video1.
Choose 3 to 30 seconds and 720p or 1080p; the exact cost shows on the Generate button. Watch the result and change one part of the prompt at a time.
A character or presenter who speaks a line, with the room's sound around them.
Up to 30 seconds in one take, for scenes that need time to unfold.
Reference images and videos to keep a product, a character or a motion consistent.
Pick the plan that matches your pace
Subscribers get +20% extra credits at the same price
Subscribers receive: 1,080
Subscribers receive: 3,600
Subscribers receive: 6,600
Subscribers receive: 14,400
Subscriptions renew until canceled. Credit packs are one-time and expire 180 days after grant. Refunds & Billing · Terms · Privacy Policy · Contact Us · Pricing
aiMakeVideo is an independent AI video platform and is not affiliated with the model providers.
Kling 4.0 is Kuaishou's latest video generation model. It creates clips of 3 to 30 seconds from text, a first frame, first and last frames, or reference images and videos, and generates sound together with the picture.
It runs on credits, priced per second by resolution, and reference videos add to the price; the exact cost appears on the Generate button before you start. A run that fails for a technical reason is refunded automatically.
Yes. Sign in with a free account; every Kling 4.0 run is billed to your credit balance.
Yes. Without a reference video, every clip comes with native audio, and there is no switch to turn it off. Describe the sounds you want and write spoken lines in quotes. When you attach a reference video, that video's own sound is not kept.
From 3 to 30 seconds, in whole seconds, at 720p or 1080p.