Direct the movement
Describe what moves and what stays still. Separate subject action from camera movement.
This mode is coming later. Try text-to-video or image-to-video with Kling 3.0.
Task mode
Create a new video from reference media
Upload Media *
Supports images, videos, and audio. Mix and match.
Aspect ratio is automatically managed by the selected model.
Resolution is automatically managed by the selected model.
Duration is automatically managed by the selected model.
Content coming soon
This model overview is still being prepared. You can still start creating right away.
Learn how visual references guide a scene. Multimodal Kling 4.0 generation is not connected here yet.
1
Describe the subject, setting, action, lighting and camera. A focused scene is easier to refine.
2
Kling 3.0 via KIE supports 3–15 seconds, 16:9, 9:16 or 1:1, and model-specific resolution and audio options.
3
Check the credit cost before submitting. Review subject consistency, motion and audio in My Creations. Change one part of your prompt at a time.
Describe what moves and what stays still. Separate subject action from camera movement.
The connected Kling 3.0 workflow accepts a starting image and an optional ending image. This differs from Kling 4.0’s announced ten-keyframe control.
Native audio can add dialogue and sound. Listen to the result and check timing before publishing.
| Capability | Kling 3.0 via KIE | Kling 4.0 official model |
|---|---|---|
| Generation length | 3–15 seconds | 3–30 seconds |
| Frame guidance | First / last frames | Up to 10 keyframes |
| Site availability | Enabled after connection checks | Not yet connected |
Official model specifications do not automatically become available through this website. Read the Kling 4.0 guide for the full announcement and its limitations.