GET

/v1/schemas/fal/video/fal-ai/kling-video/v3/4k/image-to-video

Accept:

kind inputactivity videohash b232a05d…extracted Oct 14, 2026version currentsource upstream spec ↗

field	recorded source
schema	Source linked above. upstream-spec Source trace `3de772c04c77697007dc4d8387fd229376243a2e43f02209d435beb40e2c8e96` { "fetchedAt": 1791954060, "extractorVersion": "1", "verifiedAt": null }

Request body

11 properties · 1 required · 2 $defs

property	type	description	constraints
start_image_url*	string	URL of the image to be used for the video
cfg_scale	number	The CFG (Classifier Free Guidance) scale is a measure of how close you want the model to stick to your prompt.	0 ≤ n ≤ 1 · default: 0.5
duration	enum	The duration of the generated video in seconds	3 \| 4 \| 5 \| 6 \| 7 \| 8 · +7 more · default: "5"
end_image_url	string	URL of the image to be used for the end of the video	nullable
elements	KlingV3ComboElementInput[]	Elements (characters/objects) to include in the video. Each example can either be an image set (frontal + reference images) or a video. Reference in prompt as @Element1, @Element2, etc.	nullable
shot_type	enum	The type of multi-shot video generation. 'intelligent' lets the model automatically determine shot structure.	customize \| intelligent · default: "customize"
prompt	string	Text prompt for video generation. Either prompt or multi_prompt must be provided, but not both.	length ≤ 2500 · nullable
generate_audio	boolean	Whether to generate native audio for the video. Supports Chinese and English voice output. Other languages are automatically translated to English. For English speech, use lowercase letters; for acronyms or proper nouns, use uppercase.	default: true
negative_prompt	string		length ≤ 2500 · default: "blur, distort, and low quality"
multi_prompt	KlingV3MultiPromptElement[]	List of prompts for multi-shot video generation. If provided, divides the video into multiple shots.	nullable
enable_safety_checker	boolean	Disabling safety checks requires account authorization; unauthorized requests are always checked.	default: true

Types

2 $defs

KlingV3ComboElementInputobject

reference_image_urls	string[]	Additional reference images from different angles. 1-3 images supported. At least one image is required.	nullable
frontal_image_url	string	The frontal image of the element (main view).	nullable
voice_id	string	The voice ID for this element. The voice will be binded to the element and references to this element will use the binded voice. Get voice IDs from the following endpoint: https://fal.ai/models/fal-ai/kling-video/create-voice	nullable
video_url	string	The video URL of the element. A request can only have one element with a video.	nullable

KlingV3MultiPromptElementobject

duration	enum	The duration of this shot in seconds	1 \| 2 \| 3 \| 4 \| 5 \| 6 · +9 more · default: "5"
prompt*	string	The prompt for this shot.

Validate a payload

before you spend tokens

shell

$ curl -X POST https://modelschemas.com/v1/validate \
    -d '{"provider":"fal","endpointId":"fal-ai/kling-video/v3/4k/image-to-video","payload":{…}}'
→ {"valid": false, "errors": [{"path": "#", "keyword": "required", …}]}