Kling 4.0

Overview

The Kling 4.0 image-to-video route is designed for developers who need to generate continuous video from a single first-frame image. Provide a jpeg or png image along with a description of motion, camera movement, and mood to obtain a 3- to 30-second clip, with support for 720p and 1080p resolutions and multiple aspect ratios including 16:9, 9:16, 1:1, and 21:9, billed by output seconds. It is suited to creative workflows that need to start from a fixed frame and control camera direction, such as advertising storyboards, short social videos, and narrative previsualization.

Key Highlights

• Lock the starting frame with a single image — Use one image as the shot's starting point and control motion and camera direction with a prompt, reducing the visual drift that comes from text-only generation.
• Adjustable duration and aspect ratio by scene — Duration ranges from 3 to 30 seconds, and the aspect ratio can be landscape, portrait, square, or widescreen; if unspecified, it follows the source image's ratio.
• Tiered resolution for drafts and final cuts — Use 720p for quick test clips and 1080p for delivery-grade output; the latter takes noticeably longer to render, so you can switch by stage.

Capabilities

First-frame-driven video generation: Given an image and a prompt, the model generates video; the prompt describes motion, camera movement, and atmosphere, and the model continues the shot using the first frame as the visual baseline.

3- to 30-second duration control: Specify the clip length with the duration parameter. Short clips are for quick validation, while long clips reduce the work of generating multiple times and stitching them together; billing is calculated by output seconds.

Selectable resolution and aspect ratio: Supports two output tiers, 720p and 1080p, with aspect ratios of 16:9, 9:16, 1:1, and 21:9. If left empty, the source image's ratio is used, making it easy to fit different placements in one go.

Sound-direction prompts: The prompt field can also describe the desired sound, expressing visual motion and auditory direction in the same call.

Use Cases

• Advertising storyboards: extend a shot from a fixed product frame — Input a product image and a camera-movement description to produce a dynamic clip of a few seconds to tens of seconds, for ad assets or storyboard previsualization.
• Short social videos: turn a portrait frame into motion — Use the 9:16 aspect ratio to generate a short video clip from a single composition, ready for portrait-format publishing.
• Narrative previsualization: advance the story from key frames — Generate segments from a scene's first frame with motion and mood descriptions, for low-cost validation of shot pacing and narrative continuity.
• Widescreen atmospheric shots — Use the 21:9 aspect ratio to generate an atmospheric long take from a still image, for opening titles, transitions, or background visuals.

Limitations and Selection Advice

Only a single first-frame image input is supported; multiple reference assets, keyframe sequences, and first/last-frame control are not available. The maximum duration per generation is 30 seconds, and resolution is limited to 720p and 1080p. 1080p rendering takes noticeably longer, so before delivering a final cut, it is advisable to first validate the shot and motion direction at 720p. When generating video from text alone, choose the text-to-video route in the same series; for long narratives that require precise multi-shot continuity, generate in segments and then edit them together.