Let the whole moment unfold
Give a scene room for an opening, a change, and a payoff. Plan how the setting, movement, and emotion carry through the sequence so the whole moment feels connected.
Start creatingAvailable on Vuppo
Explore the next chapter of AI video with Kling 4.0. Plan complete scenes, shape the moments that matter, and bring characters, movement, and dialogue into one creative direction.
Create a scene from text, animate a still image, or guide the subject and movement with image and video references.
Give a scene room for an opening, a change, and a payoff. Plan how the setting, movement, and emotion carry through the sequence so the whole moment feels connected.
Start creatingPlan where the subject begins, when the reveal happens, and how the shot ends. Clear visual milestones give the scene a direction before you generate.
Start creatingGive your subject, setting, and visual style a shared direction. Assign each reference a purpose so recurring characters and the world around them feel connected.
Start creatingDecide who speaks, what they say, and how the line should feel. Build the dialogue around the character’s intention so the voice belongs to the scene.
Start creatingUse this three-step workflow to turn a prompt, image, or references into a directed Kling 4.0 clip.
Give every scene a purpose, from a dramatic reveal to a memorable product moment. Find a creative direction for the story you want to tell.
Build a scene around one decision, a reaction, or a reveal. Keep the cast and setting focused so the audience can follow the emotional change from opening to final frame.
Start creatingShow the product in use, then give its most distinctive detail a moment of its own. Plan the light, background, and final composition around the benefit you want people to remember.
Start creatingChoose a visual motif and let it evolve with the music. Alternate performance, movement, and atmosphere while keeping the colors and character styling connected.
Start creatingOpen with one image or action that makes people stop scrolling. Reuse a character, setting, or color palette across short episodes so each post feels part of the same series.
Start creatingBrowse creator prompts with matching video examples, then copy a prompt to adapt it for your next scene.
Explore Kling 4.0 video promptsKling 4.0 is an AI video model connected through Fal. Create from text alone, animate a start image with an optional end frame, or use image and video references. It supports 3–30 second clips at 720p or 1080p.
Kling 4.0 and Kling 4.0 Flash are anticipated variants of the next model family. The key things to compare will be generation speed, visual quality, creative controls, and cost. Their final differences will become clear with the release.
Kling 4.0 supports 3–30 second clips through the connected Fal text-to-video, image-to-video, and reference-to-video endpoints.
Compared with Kling 3.0, Kling 4.0 is expected to offer more scope for complete scenes, visual planning, reference control, and character voices. Final specifications and direct testing will determine how much those capabilities change in practice.
The connected endpoint supports 720p and 1080p output. 4K is not offered in this workflow.
Multilingual character dialogue is an expected direction for Kling 4.0; the supported languages and accents will depend on the final release. Write the language, speaker, and delivery beside each line so your intent is clear.
Reference mode accepts up to 10 images and 5 videos. Mention them as @Image1 or @Video1 in the prompt; Frames mode uses a start image and optional end frame. Leave media empty in either mode to generate from text only.
The current integration supports text-to-video, image-to-video, and reference-to-video generation. Use the existing video editor after generating a clip when you need precise cuts or captions.
Keep in-scene wording short and describe its position, size, and contrast clearly. Kling 4.0’s text accuracy has not been verified here, so add essential names, prices, and captions in editing when every character must be exact.