Video models

Kling O3

On this page

Operations
Text To Video — kling-o3-text-to-video
Image To Video — kling-o3-image-to-video
Transformation — kling-o3-transformation
Reference To Video — kling-o3-reference-to-video

Kling 3.0 Omni — cinematic video with native sound, start/end frames and multi-shot prompting.

modelId	kling-o3-text-to-video, kling-o3-image-to-video, kling-o3-transformation, kling-o3-reference-to-video
Modality	video
Pricing	See this model on the Pricing page for the current per-call price (with your markup).

Operations

Operation	modelId	Endpoint	Required input
Text To Video	kling-o3-text-to-video	POST /api/v1/generate	prompt, customize_multi_shots, duration, resolution
Image To Video	kling-o3-image-to-video	POST /api/v1/generate	prompt, customize_multi_shots, duration, resolution
Transformation	kling-o3-transformation	POST /api/v1/generate	prompt, video_urls, resolution
Reference To Video	kling-o3-reference-to-video	POST /api/v1/generate	prompt, customize_multi_shots, duration, resolution
Poll task	—	GET /api/v1/task/{id}?model=kling-o3-text-to-video	—

Text To Video — kling-o3-text-to-video

Field	Type	Required	Values / example
prompt	string	Yes	Text value Max 2500 characters. (example: Create a 15-second photorealistic cinematic boxing drama, 16:9, emotionally intense, like an epic life-drama sports film. Inside a packed old boxing arena… — full value in the request example)
prefer_multi_shots	boolean	No	Smart storyboarding — let the model split the clip into shots. (true/false) (default: false)
customize_multi_shots	boolean	Yes	Generate multi-camera video. (true/false) (default: false)
multi_prompt	object[]	No	Up to 6 shots, each 1-12 seconds. Their durations must add up to Duration. (array of { prompt, duration })
audio	boolean	No	The video contain audio (true/false) (default: true)
duration	number	Yes	Range 3-15 (default: 5)
resolution	string	Yes	Options: 720p \| 1080p \| 4K (default: 720p)
aspect_ratio	string	No	Options: 16:9 \| 9:16 \| 1:1 (default: 16:9)

bash

curl -X POST https://you.bot/api/v1/generate \
  -H "Authorization: Bearer $YOUBOT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"modelId":"kling-o3-text-to-video","input":{"prefer_multi_shots":false,"customize_multi_shots":false,"audio":true,"duration":5,"resolution":"720p","aspect_ratio":"16:9"}}'

Image To Video — kling-o3-image-to-video

Field	Type	Required	Values / example
image_urls	string[]	No	Start Frame / End Frame
prompt	string	Yes	Text value Max 2500 characters. (example: A paper boat drifts down a rain-swollen gutter at night, city lights breaking on the water, slow push-in.)
prefer_multi_shots	boolean	No	Smart storyboarding — let the model split the clip into shots. (true/false) (default: false)
customize_multi_shots	boolean	Yes	Generate multi-camera video. (true/false) (default: false)
multi_prompt	object[]	No	Up to 6 shots, each 1-12 seconds. Their durations must add up to Duration. (array of { prompt, duration })
audio	boolean	No	The video contain audio (true/false) (default: true)
duration	number	Yes	Range 3-15 (default: 5)
resolution	string	Yes	Options: 720p \| 1080p \| 4K (default: 720p)
aspect_ratio	string	No	Options: 16:9 \| 9:16 \| 1:1 (default: 16:9)

bash

curl -X POST https://you.bot/api/v1/generate \
  -H "Authorization: Bearer $YOUBOT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"modelId":"kling-o3-image-to-video","input":{"image_urls":["https://you.bot/examples/grok-imagine-text-to-image.jpg"],"prefer_multi_shots":false,"customize_multi_shots":false,"audio":true,"duration":5,"resolution":"720p","aspect_ratio":"16:9"}}'

Transformation — kling-o3-transformation

Field	Type	Required	Values / example
prompt	string	Yes	Text value Max 2500 characters. (example: Transform the daytime scene into night, keeping the motion and framing.)
video_urls	string[]	Yes	The source video to transform. Required for this mode. (image URL)
image_urls	string[]	No	Up to 4 reference images — the subject, style or scene to carry into the video. (image URL)
audio	boolean	No	The video contain audio (true/false) (default: true)
resolution	string	Yes	Options: 720p \| 1080p \| 4K (default: 720p)

bash

curl -X POST https://you.bot/api/v1/generate \
  -H "Authorization: Bearer $YOUBOT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"modelId":"kling-o3-transformation","input":{"prompt":"Transform the daytime scene into night, keeping the motion and framing.","video_urls":["https://you.bot/examples/gemini-omni.mp4"],"image_urls":["https://you.bot/examples/grok-imagine-text-to-image.jpg"],"audio":true,"resolution":"720p"}}'

Reference To Video — kling-o3-reference-to-video

Field	Type	Required	Values / example
image_urls	string[]	No	Up to 4 reference images — the subject, style or scene to carry into the video. (image URL)
video_urls	string[]	No	The source video to transform. Required for this mode. (image URL)
prompt	string	Yes	Text value Max 2500 characters. (example: A paper boat drifts down a rain-swollen gutter at night, city lights breaking on the water, slow push-in.)
prefer_multi_shots	boolean	No	Smart storyboarding — let the model split the clip into shots. (true/false) (default: false)
customize_multi_shots	boolean	Yes	Generate multi-camera video. (true/false) (default: false)
multi_prompt	object[]	No	Up to 6 shots, each 1-12 seconds. Their durations must add up to Duration. (array of { prompt, duration })
audio	boolean	No	The video contain audio (true/false) (default: true)
duration	number	Yes	Range 3-15 (default: 5)
resolution	string	Yes	Options: 720p \| 1080p \| 4K (default: 720p)
aspect_ratio	string	No	Options: 16:9 \| 9:16 \| 1:1 (default: 16:9)

bash

curl -X POST https://you.bot/api/v1/generate \
  -H "Authorization: Bearer $YOUBOT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"modelId":"kling-o3-reference-to-video","input":{"image_urls":["https://you.bot/examples/grok-imagine-text-to-image.jpg"],"video_urls":["https://you.bot/examples/gemini-omni.mp4"],"prefer_multi_shots":false,"customize_multi_shots":false,"audio":true,"duration":5,"resolution":"720p","aspect_ratio":"16:9"}}'

Kling 3.0 Flux.1 Kontext