Kling 4.0 AI Video Generator

Make videos up to 30 seconds from text, images or your own clips, with sound generated in the same pass.

Kling 4.0 is the newest Kling video model from Kuaishou, built for longer clips and for following footage you already have.

Text, image & video input
Up to 30 seconds
Native audio
11M+ users

Kling 4.0 Specs at a Glance

This clip is one 30-second Kling 4.0 generation from a single prompt: four scenes, the camera moves and the studio sound, with no editing afterwards. Sound on.

Text to video Generate a clip from a written prompt alone, with no image needed.
First and last frame Set the opening shot, the closing shot, or both ends of the clip.
Up to 10 reference images Show Kling the people, products and places that should appear.
Up to 5 video references Follow the motion, timing and camera of clips up to 30 seconds combined.
Native audio Dialogue, ambience, effects and music generated together with the picture.
3 to 30 seconds, up to 1080p Any length from 3 to 20 seconds, plus 25 and 30, in 1:1, 9:16, 16:9 or 21:9.

Kling 4.0

Copy the motion from a clip you already have

Attach up to 5 video clips, up to 30 seconds in total, and Kling 4.0 follows their movement, timing and camera work in a new scene. Add up to 10 reference images in the same generation to lock in a character, a product or a location, then describe the rest in your prompt. Clips run from 3 to 30 seconds in 720p or 1080p, with sound generated alongside the picture.

Prompting with Video References

With a clip attached, the prompt no longer has to describe the motion. Split the work between the clip, the images and the text, and give each one a single job.

What the clip sets

Movement, timing and camera work: a dance, a product spin, a walk across a room, a drone pass. Tell Kling to keep the movement and camera from the reference video instead of describing them again.

First and last frame

Set where the clip ends

Upload an image as the last frame and the clip ends on it. Use it to cut cleanly into your next shot, line up a transition or make a loop that returns to where it started. Kling 4.0, Kling O3 and Kling 3.0 Pro all take a first and a last frame.

kling 2.1 pro last frame control

Image to Video

Animate a photo or product shot

Upload an image as the first frame and describe what happens next. A product can turn on its base, a portrait can get a slow camera move around the subject. The clip starts from your picture, so the composition and colors are the ones you chose.

kling image to video ai online generator

How to use Kling 4.0 on getimg.ai

Your first Kling 4.0 clip takes three steps.

1. Open the Create video generator

Click here and select Kling 4.0 from the model picker in the prompt box. Kling O3, 3.0 Pro and 3.0 Turbo are in the same list.

2. Write a prompt

Describe the subject, action, setting, camera and sound. If you like, add a first or last frame, reference images or video clips to follow.

3. Generate

Pick 16:9, 9:16, 1:1 or 21:9, a length from 3 to 30 seconds, and 720p or 1080p. Video and sound come back together in one clip.

Night street food stall in Bangkok. A cook in a white t-shirt tosses rice noodles in a large wok as flames leap up from the burner. Medium shot at counter height, the camera slowly arcs to the left around the stall, steam and smoke catching the orange light of hanging bulbs. Audio: roaring gas burner, a metal spatula scraping the wok, sizzling oil, distant motorbikes and street chatter, no music.

Video
16:9
5s

How to Prompt Kling 4.0

A dependable Kling 4.0 prompt runs in one order: subject, action, setting, camera, lighting, sound. You don't need all six every time, but keeping the order makes the important parts hard to miss.

Give each shot one main action. Several competing actions in one clip are where motion tends to break, and the risk grows at 25 and 30 seconds.

Name the camera move: tracking shot, push-in, orbit, handheld. A named move gives Kling something concrete to follow, where "cinematic" on its own doesn't.

Short prompts work too. Prompt enhancement fills in detail when you start from a single line.

A cyclist in a red rain jacket rides down a narrow cobbled street in Lisbon at dusk. She stands on the pedals as the road tilts uphill. Low tracking shot from the side, slight handheld movement, warm shop lights reflecting on wet stone. Tyres on cobbles, a distant tram bell, light rain, no music.

Video
16:9
5s

Native Audio, Dialogue Included

Kling 4.0 generates sound in the same pass as the picture. Write it into the prompt like the rest of the shot: spoken lines in quotes and who says them, the background noise, the effects, and whether there's music. If you leave the sound out, Kling chooses it for you.

Kling 4.0 doesn't take audio files as references, so a voice or a music track has to be described in words.

Foggy harbor dock in Maine in the early morning. A weathered fisherman in his sixties wearing a yellow oilskin jacket and a knit cap sits on a wooden crate mending a green fishing net. Medium shot with his whole head and shoulders in view, static camera with a slow push-in. He looks up toward someone off camera and says in a calm, gravelly voice: "Fog's lifting by noon. We'll head out after lunch." Then he goes back to the net. Audio: gulls, water lapping against the pilings, a distant boat engine, his voice, no music.

Video
16:9
5s

Kling 4.0, O3, 3.0 Pro or 3.0 Turbo?

The Kling models on getimg.ai trade length, resolution and control. We ran the same prompt through all four at 5 seconds, 16:9 and 720p, so you can compare motion and detail yourself.

Top left: Kling 4.0. Top right: Kling O3. Bottom left: Kling 3.0 Pro. Bottom right: Kling 3.0 Turbo.

Kling 2.6 Pro and Kling 2.5 Pro Turbo are also in the model picker for 5 and 10 second clips at 1080p.

Kling 4.0

3 to 30 seconds. 720p and 1080p. 1:1, 9:16, 16:9 and 21:9. First and last frame, up to 10 reference images, up to 5 video references. Native audio. Best for longer clips and following existing footage.

Kling O3

4 to 10 seconds. Up to 4K. 16:9 and 9:16. First and last frame, reference images. Native audio. Best for short, high-resolution shots with a planned ending.

Kling 3.0 Pro

4 to 15 seconds. Up to 4K. 16:9 and 9:16. First and last frame. Native audio. Best for 4K clips longer than 10 seconds.

Kling 3.0 Turbo

4 to 15 seconds. 720p and 1080p. 1:1, 16:9 and 9:16. First frame. Native audio. Best for quick drafts before a final run.

Why Run Kling on getimg.ai

One account for Kling and the other image and video models you use, in the browser.

No separate Kling account

Kling runs inside the getimg.ai Create video generator. Sign in, pick a Kling model and generate, with no API keys or extra subscriptions.

Next to your other models

Make a first frame with an image model, animate it with Kling, then upscale the result, all in the same app.

Included in your plan

Kling runs on the same credits as every other image and video model, so there's no separate Kling subscription to manage.

Reuse what you make

Download your clip as an MP4, or attach it as a video reference for your next Kling 4.0 generation.

Frequently Asked Questions

Make your first Kling 4.0 clip

Start from a prompt, a photo or a clip you already have, and get up to 30 seconds of video with sound.