API token from https://platform.acedata.cloud
Specifies the format of the response from the server, should be application/json.
application/json The action to generate the video. If text2video, the video will be generated from the prompt.
text2video, image2video, extend Mode of generating video, optional, enumeration value: std (High performance) or pro (High performance). Default value is std.
std, pro The model used to generate the videos, default value is kling-v1
kling-v1, kling-v1-6, kling-v2-master, kling-v2-1-master, kling-v2-5-turbo, kling-video-o1 The prompt which used to generate the video.
"White ceramic coffee mug on glossy marble countertop with morning window light. Camera slowly rotates 360 degrees around the mug, pausing briefly at the handle."
Video generation duration, in seconds, optional, enumeration value: 5,10 (Requests that include endframes only support 5 seconds).
5, 10 When the action is extend, you can specify video_id to extend the video.
Degree of freedom to generate video, optional. The greater the value, the stronger the correlation. Value range: [0,1].
The aspect ratio of the video, optional, enumeration value: 16:9, 9:16, 1:1. Default value is 16:9.
16:9, 9:16, 1:1 The callback URL which will be called when the video is generated
You can specify the end frame to generate video, when the action is image2video.
Contains 6 Fields, used to specify the camera's movement or change in different directions. camera_control.type should be one of 'simple', 'down_back', 'forward_up', 'left_turn_forward', 'right_turn_forward'. camera_control.config is an optional object, contains 6 optional fields - camera_control.config.horizontal, camera_control.config.vertical, camera_control.config.pan, camera_control.config.tilt, camera_control.config.roll, camera_control.config.zoom. The value of each field is a number, and the value range is [-1,1].
This parameter is used to specify a list of reference subjects when generating content, configured in the element_list array using element_id (the ID from the subject library). The number of reference subjects must be counted together with the number of reference images under the following limits: if there is a reference video, reference subjects + reference images ≤ 4; if there is no reference video, reference subjects + reference images ≤ 7.
This parameter is used to upload a reference video via video_url (MP4/MOV format, duration between 3–10 seconds, resolution between 720px and 2160px, frame rate between 24–60fps, file size ≤200MB, and only 1 video allowed). You can set the refer_type to feature (feature reference video) or base (video to be edited, default); videos set as base cannot define start or end frames. You can use keep_original_sound to choose whether to retain the original audio (yes to keep, no to remove, applicable to both types). The video_url must not be empty.
The value is optional and contains a maximum of 200 characters.
You can specify the first frame to generate video, when the action is image2video.