text to video | single-shot-video generation | ✅ | ✅ |
multi-shot-video generation | ✅ | ✅ | |
voice control | ❌ | ❌ | |
others | - | - | |
image to video | single-shot-video generation （only start frame） | ✅ | ✅ |
multi-shot-video generation | ✅ | ✅ | |
start & end frame | ✅ | ✅ | |
element control （video character elements & multi-image elements） | ✅ | ✅ | |
motion control | （coming soon） | （coming soon） | |
voice control | ❌ | ❌ | |
others | - | - |
Image Generation
kling-v3-omni | custom aspect ratio（1K/2K/4K） | intelligent aspect ratio | |
text to image | single-image generation | ✅ | ✅ |
others | - | - | |
image to image | single-image generation | ✅ | ✅ |
series-image generation | ✅ | ✅ | |
element control （only multi-image elements） | ✅ | ✅ | |
others | - | - | |
kling-image-o1 | custom aspect ratio（1K/2K） | intelligent aspect ratio | |
text to image | single-image generation | ✅ | - |
others | - | - | |
image to image | single-image generation | ✅ | ✅ |
element control （only multi-image elements） | ✅ | ✅ | |
others | - | - | |
kling-v3 | custom aspect ratio（1K/2K） | intelligent aspect ratio | |
text to image | single-image generation | ✅ | - |
others | - | - | |
image to image | single-image generation | ✅ | - |
element control （only multi-image elements） | ✅ | - | |
others | - | - | |
API
Omni-Video - Create Task
Protocol | https |
Request URL | /v1/videos/omni-video |
Request Method | POST |
Request Format | application/json |
Response Format | application/json |
Request Header
Field | Value | Description |
Content-Type | application/json | Data Exchange Format |
Authorization | Authentication information, refer to API authentication | Authentication information, refer to API authentication |
Request Body
Field | Type | Required Field | Default Value | Description |
model_name | string | Optional | kling-video-o1 | Model Name ●Enum values：kling-video-o1, kling-v3-omni |
multi_shot | boolea n | Optional | false | Whether to generate multi-shot video ●true: the prompt parameter is invalid ●false: the shot_type parameter and the multi_prompt parameter are invalid |
shot_type | string | Optional | None | Storyboard method ●Enum values：customize ●When the multi_shot parameter is set to true, the current parameter is required. |
prompt | string | Optional | None | Text prompt words, which can include positive and negative descriptions ●The prompt words can be templated to meet different video generation needs. 📌 The Omni model can achieve various capabilities through Prompt with elements, images, videos, and other content. 1.Specify a element, image, or video in the format of<<<>>, such as<<element_1>>>,<<<image_1>>>,<<<video_1>>>. 2.For more information, please refer to: Kling VIDEO 3.0 Omni Model User Guide ●Cannot exceed 2500 characters. ●When the multi_shot parameter is false or the shot-type parameter is intelligence, the current parameter must not be empty. Different model versions support different scopes. For details, please refer to the capability map mentioned above. |
multi_prompt | array | Optional | None | Information about each storyboard, such as prompts and duration ●Define the shot sequence number, corresponding prompt word, and duration through the index, prompt, and duration parameters, where: ○Supports up to 6 storyboards, with a minimum of 1 storyboard. ○The maximum length of the content for each storyboard 512. ○The duration of each storyboard should not exceed the total duration of the current task, and it must not be less than 1. ○The sum of the durations of all storyboards equals the total duration of the current task. ●Load with key:value, details as follows: ●When the mult_shot parameter is set to true and the shot_type parameter is set to customize, the current parameter must not be empty. |
image_list | array | Optional | None | Reference Image List ●Including reference images of the element, scene, style, etc., it can also be used as the start or end frame to generate videos; When generating a video as the start or end frame: ○Define whether the image is in the first and last frames using the type parameter: first_frame is the start frame, end_frame is the end frame. ■Currently does not support only the end frame, which means that when there is a end frame image, there must be a first frame image. ○When generating a video using the first frame or the first and last frames, video editing functions cannot be used. ●Load with key:value, details as follows: ●Supports inputting image Base64 encoding or image URL (ensure accessibility). ●Supported image formats include.jpg / .jpeg / .png. ●The image file size cannot exceed 10MB, and the width and height dimensions of the image shall not be less than 300px, and the aspect ratio of the image should be between 1:2.5 ~ 2.5:1. ●The amount of reference image is related to whether there is refrence video and the amount of reference element: ○When there are reference video, the sum of the amount of reference image and the amount of reference element shall not exceed 4. ○When there is no reference video, the sum of the amount of reference image and the amount of reference element shall not exceed 7. ○Setting an end frame is not supported when there are more than 2 images. ●The value of image_url parameter must not be empty. Different model versions support different scopes. For details, please refer to the capability map mentioned above. |
element_list | array | Optional | None | Reference Element List ●Based on element ID configuration. Load with key:value, details as follows: ●The amount of reference element is related to whether there is reference video and the amount of reference image: ○ When using raw video from the first frame or raw video from the first and last frames, a maximum of 3 subjects is supported. ○ When there is a reference video, the sum of the number of reference images and the number of reference subjects must not exceed 4, and the use of video subjects is not supported. ○ When there is no reference video, the sum of the number of reference images and the number of reference subjects must not exceed 7. ●The elements are categorized into video customization element (named as video character elements) and image customization elements (named as Multi-Image Elements), each with distinct scopes of application. Please exercise caution in distinguishing between them. ●For more detailed information on the subject, please refer to: Kling Element Library User Different model versions support different scopes. For details, please refer to the capability map mentioned above. |
video_list | array | Optional | None | Reference Video, get link for uploaded video. ●It can be used as a reference video for feature or as a video to be edited, with the default being the video to be edited; Selective retention of video original sound. ○Distinguish reference video types based on the refer_type parameter: feature is the feature reference video, base is the video to be edited. ○When the reference video is a video to be edited, the start and end frames of the video cannot be defined. ○Select whether to keep the video original sound through the parameter keep_original_stound, with yes indicating retention and no indicating non retention; The current parameters also apply to the feature reference video. ●When there is a reference video, the value of the sound parameter can only be off. ●Load with key:value, details as follows: ●Only .mp4/.mov formats are supported. ●Only supports videos with a duration of ≥ 3 seconds and ≤ 10 seconds. ●Video resolution must be between 720px and 2160px (inclusive) in both width and height. ●Only supports videos with frame rates of 24 ~ 60 fps, the output result is 24 fps. ●Only supports 1 video can be uploaded, with a video size not exceeding 200MB. ●The value of video_url parameter must not be empty. Different model versions support different scopes. For details, please refer to the capability map mentioned above. |
sound | string | Optional | off | Is sound generated simultaneously when generating videos ●Enum values: on, off Only V2.6 and subsequent versions of the model supports the current parameters |
mode | string | Optional | pro | Video generation mode ●Enum values: std, pro ●std: Standard Mode, generating 720P videos, which is cost-effective ●pro: Professional Mode, generating 1080P videos, which is higher quality video output Different model versions support different scopes. For details, please refer to the capability map mentioned above. |
aspect_ratio | string | Optional | None | The aspect ratio of the generated video frame (width:height) ●Enum values：16:9, 9:16, 1:1 ●This parameter is required when the first-frame reference or video editing features are not used. |
duration | string | Optional | 5 | Video Length, unit: s (seconds) ●Enum values: 3，4，5，6，7，8，9，10，11，12，13，14，15: ○When using the video editing function ("refer_date": "base"), the output result is the same as the duration of the incoming video, and the current parameter is invalid. Calculate billing by rounding the input video duration to the nearest integer. Different model versions support different scopes. For details, please refer to the capability map mentioned above. |