Kling 4.0 AI Video Generator & Editor for Scene-by-Scene Storytelling

Generate full 30-second scenes with Kling 4.0, then edit and extend them on a Pictell canvas. Connect character, product, motion, and voice references, swap subjects or backgrounds, and chain clips so every scene picks up where the last one ended.

Kling 4.0AI Video Generator & Editor for Scene-by-Scene Storytelling

How to build a scene with Kling 4.0

Start from the sample canvas, then grow it into a sequence of connected scenes

Open the sample edit

Create a project from the canvas above. It already wires a prompt, a character image, and a source video into Kling V4.0 Reference — a ready-made video edit you can run or change.

Connect references for consistency

Attach up to 10 images, 5 videos, and 3 audio clips. Call them in the prompt as @Image1, @Video1, and @Audio1 to lock a character's look, borrow a camera move, or keep a voice.

Edit, extend, and reuse

Set 3–30 seconds at 720p or 1080p and generate. Feed the result into the next node to restyle it, replace a subject, or carry its motion into the next scene. Save the canvas and the whole chain is reusable.

30 seconds holds a whole scene

30 seconds holds a whole scene

Kling 4.0 generates 3 to 30 seconds in a single pass. An entrance, a line of dialogue, an action, and a reaction can all live in one clip, so you plan in complete scenes instead of 5-second fragments.

Big motion and complex camera work

Big motion and complex camera work

According to Kuaishou, Kling 4.0 keeps body structure and spatial layout more stable through fast action, tracking shots, and orbiting cameras, with more natural expressions, lighting, and lip sync. Fight, chase, and dance scenes that used to fall apart are now worth trying.

Edit existing video with references

Edit existing video with references

Mix image, video, and audio references in one generation: an image sets the character's appearance, a video sets the camera and timing, and audio keeps the voice. Plug an existing clip back in to replace the background, change the style, or swap the subject while keeping the original pacing.

From single clips to a connected story

From single clips to a connected story

Every prompt, reference, and result stays on the canvas as a node. Branch one scene into variations, reuse the same character references across shots, and line scenes up into a longer piece — with a visible record of how each scene was made.

Frequently Asked Questions

What is Kling 4.0?
Kling 4.0 is Kuaishou's latest AI video model, officially released on October 15, 2026. Compared with earlier versions, it is more stable in large motion and complex camera moves, natively generates up to 30 seconds, and accepts mixed image, video, and audio references.
Why use Kling 4.0 on a canvas instead of a single prompt box?
Longer videos are made of several scenes that need the same characters, products, and voices throughout. On a Pictell canvas, references and finished clips are nodes you can connect again, so each scene builds on the last instead of starting from zero.
Which Kling 4.0 modes are available on Pictell?
Four: Kling V4.0 for text-to-video, Kling V4.0 Image for image-to-video with a first frame, optional last frame and in-between keyframes, Kling V4.0 Reference, and Kling V4.0 Edit. The sample canvas uses Reference; you can switch models inside the video node.
Can I edit an existing video with Kling 4.0?
Yes. Kling V4.0 Edit rewrites a source video from a text instruction while keeping its motion. Add up to 10 reference images and refer to them as @Video1 and @Image1. Length follows the source clip, up to 30 seconds. Original sound is kept by default, or you can generate new synced audio. Use Kling V4.0 Reference when you want a new scene that only borrows the camera or performance.
How many reference files can I use?
Kling V4.0 Reference accepts up to 10 images, 5 videos, and 3 audio clips per generation. Refer to them in the prompt as @Image1, @Video1, and @Audio1.
How long are clips, and what quality is supported?
Each generation runs 3 to 30 seconds at 30 fps, in 720p or 1080p. Aspect ratios are 16:9, 9:16, 1:1, and 21:9, and Reference mode also supports auto.
Does Kling 4.0 generate audio?
Yes. Text-to-video and image-to-video always return synced stereo audio. Video edit keeps the original sound by default, or can generate new synced audio. In Reference mode, you can keep or mute the reference video's sound.
Does Pictell support Kling 4.0 keyframes or 4K?
Image-to-video accepts a first frame, an optional last frame, and up to 8 in-between keyframes, 10 images in total. Output is 720p or 1080p. 4K and 10-bit HDR are not available.
Do I need an account to generate?
You can preview the canvas on this page. Creating a project and generating video requires a Pictell account with available credits.

Turn Kling 4.0 clips into a story

Generate a 30-second scene, edit it with references, and build the next one on the same canvas.