← Back to all generators

kwaivgi/kling-o1

Modify an existing video through natural-language commands, changing subjects, environments, and visual style while preserving the original motion and timing.

Capabilities

Reference Images

Cost

Community model (estimated from hardware time)

Input Parameters

prompt required string

Text prompt for video generation. Can include references like <<<image_1>>>, <<<video_1>>> to reference inputs.

aspect_ratio string

Aspect ratio of the generated video. Required for text-to-video. Ignored when using first frame image or video editing.

Default: "16:9"
16:9 9:16 1:1
duration integer

Video duration in seconds. For text/image-to-video: 5 or 10. With video reference (feature type): 3-10. Ignored for video editing (base type).

Default: 5
3 4 5 6 7 8 9 10
end_image string

Last frame image for the video. Requires start_image to be set. Supports .jpg/.jpeg/.png, max 10MB.

keep_original_sound boolean

Whether to keep the original sound from the reference video.

Default: true
mode string

Video generation mode. 'std' is cost-effective, 'pro' has higher quality.

Default: "pro"
std pro
reference_images array

Reference images for elements, scenes, or styles (up to 7 without video, 4 with video). Supports .jpg/.jpeg/.png.

reference_video string

Reference video for style, camera movement, or as base for editing. Supports .mp4/.mov, 3-10s duration, max 200MB.

start_image string

First frame image for the video. Supports .jpg/.jpeg/.png, max 10MB.

video_reference_type string

How to use the reference video: 'feature' for style/camera reference, 'base' for video editing.

Default: "feature"
feature base
Version: 6d5f2d4becc7 Updated: 6/8/2026 2.1K runs