kwaivgi/kling-v3-omni-video

OfficialView on Replicate →

Kling Video 3.0 Omni: Unified multimodal video generation with reference images, video editing, native audio, and multi-shot control

Capabilities

Reference Images

Cost

Community model (estimated from hardware time)

Input Parameters

Name	Type	Description	Default	Constraints
`prompt`*	string	Text prompt for video generation. Supports <<<image_1>>>, <<<video_1>>> template references. Max 2500 characters.	`—`	—
`aspect_ratio`	string	Aspect ratio. Required when not using start frame or video editing.	`"16:9"`	16:99:161:1
`duration`	integer	Video duration in seconds (3-15). Ignored for video editing (base).	`5`	min: 3, max: 15
`end_image`	string(uri)	Last frame image. Requires start_image. Supports .jpg/.jpeg/.png, max 10MB, min 300px.	`—`	—
`generate_audio`	boolean	Generate native audio. Mutually exclusive with reference video.	`false`	—
`keep_original_sound`	boolean	Keep original sound from reference video.	`true`	—
`mode`	string	'standard' generates 720p, 'pro' generates 1080p, '4k' generates 4K. 4K does not support reference_video.	`"pro"`	standardpro4k
`multi_prompt`	string	JSON array of shot definitions for multi-shot mode. Each shot: {"prompt": "...", "duration": N}. Max 6 shots, min duration 1s per shot, total must equal duration. Example: [{"prompt":"A cat jumps","duration":3},{"prompt":"It lands","duration":2}]	`—`	—
`reference_images`	array	Reference images for elements, scenes, or styles. Supports .jpg/.jpeg/.png. Max 7 without video, 4 with video.	`—`	—
`reference_video`	string(uri)	Reference video (.mp4/.mov). Duration 3-10s, resolution 720-2160px per side, max 200MB.	`—`	—
`start_image`	string(uri)	First frame image. Supports .jpg/.jpeg/.png, max 10MB, min 300px, aspect ratio 1:2.5 to 2.5:1.	`—`	—
`video_reference_type`	string	How to use reference video: 'feature' for style/camera reference, 'base' for video editing.	`"feature"`	featurebase

promptrequiredstring

Text prompt for video generation. Supports <<<image_1>>>, <<<video_1>>> template references. Max 2500 characters.

aspect_ratiostring

Aspect ratio. Required when not using start frame or video editing.

Default: "16:9"

16:99:161:1

durationinteger

Video duration in seconds (3-15). Ignored for video editing (base).

Default: 5min: 3, max: 15

end_imagestring

Last frame image. Requires start_image. Supports .jpg/.jpeg/.png, max 10MB, min 300px.

generate_audioboolean

Generate native audio. Mutually exclusive with reference video.

Default: false

keep_original_soundboolean

Keep original sound from reference video.

Default: true

modestring

'standard' generates 720p, 'pro' generates 1080p, '4k' generates 4K. 4K does not support reference_video.

Default: "pro"

standardpro4k

multi_promptstring

JSON array of shot definitions for multi-shot mode. Each shot: {"prompt": "...", "duration": N}. Max 6 shots, min duration 1s per shot, total must equal duration. Example: [{"prompt":"A cat jumps","duration":3},{"prompt":"It lands","duration":2}]

reference_imagesarray

Reference images for elements, scenes, or styles. Supports .jpg/.jpeg/.png. Max 7 without video, 4 with video.

reference_videostring

Reference video (.mp4/.mov). Duration 3-10s, resolution 720-2160px per side, max 200MB.

start_imagestring

First frame image. Supports .jpg/.jpeg/.png, max 10MB, min 300px, aspect ratio 1:2.5 to 2.5:1.

video_reference_typestring

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

Default: "feature"

featurebase

Version: 460d4f46adf3Updated: 7/25/2026535.0K runs