Kling v3 Omni 비디오 생성
Generate videos with Kling v3 Omni using official Omni references: image_list, video_list, and element_list
중국 본토 사용자 안내: 중국 본토 사용자는 https://toapis.cn 를 API 엔드포인트(Base URL)로 사용해 주세요. 본 문서의 예시에서 https://toapis.com 을 https://toapis.cn 로 바꿔 주세요.
- Async task API, returns a task ID after submission
- Uses the official Omni reference structures:
image_list,video_list, andelement_list mode=stdmaps to 720P,mode=promaps to 1080Paudio=truegenerates an audio video and is billed as Sound- Requests with
video_listare billed as Video audioandvideo_listare mutually exclusive
Use publicly accessible image or video URLs. Do not pass base64 media data. Upload local images with the Upload 이미지 API first.
Authorization
Authorizationstring헤더필수All endpoints require Bearer Token authentication.
Authorization: Bearer YOUR_API_KEY요청 파라미터
modelstring본문필수Video generation model name, fixed as kling-v3-omni.
promptstring본문필수Text prompt. Use official Omni placeholders to reference input assets:
<<<image_N>>>referencesmetadata.image_list<<<video_N>>>referencesvideo_list<<<element_N>>>referencesmetadata.element_list
Example: "Make the person in <<<image_1>>> wave at the camera"
List order must match the placeholder order in the prompt. The system does not auto-prepend a first frame or insert placeholders for you.
modestring본문기본값: stdGeneration mode. This also determines the billing resolution.
std- standard mode, 720Ppro- professional mode, 1080P
durationinteger본문기본값: 5Video duration in seconds. Options: 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15
aspect_ratiostring본문기본값: 16:9Video aspect ratio. Common values: 16:9, 9:16, 1:1
audioboolean본문기본값: falseWhether to generate an audio video.
audio and video_list are mutually exclusive. Do not pass audio=true when video_list is provided.
video_listobject[]본문Official Omni reference video list. Reference videos in the prompt with <<<video_1>>>, <<<video_2>>>, and so on.
Show video_list object fields
video_urlstring본문필수Reference video URL. It must be publicly accessible.
refer_typestring본문기본값: baseReference type: base for source video editing, or feature for feature/motion reference.
keep_original_soundstring본문기본값: noWhether to keep the original video sound: yes or no.
metadataobject본문Extended parameters.
Show metadata fields
image_listobject[]본문Official Omni image list. Reference images in the prompt with <<<image_1>>>, <<<image_2>>>, and so on.
Show image_list object fields
image_urlstring본문필수Image URL. It must be publicly accessible.
typestring본문Image type. For first/end frame semantics, use first_frame or end_frame. If end_frame is provided, first_frame is required too.
element_listobject[]본문Official Omni subject/role reference list. Reference elements in the prompt with <<<element_1>>>, <<<element_2>>>, and so on.
Show element_list object fields
urlstring본문필수Subject image, character asset, or another official supported asset URL.
typestring본문Element type, such as image or video.
rolestring본문Role semantics supported by the official model.
watermarkboolean본문Whether to add watermark.
Omni Reference Syntax
| Syntax | 설명 |
|---|---|
<<<image_1>>> | References the first item in metadata.image_list |
<<<video_1>>> | References the first item in video_list |
<<<element_1>>> | References the first item in metadata.element_list |
The order of image_list, video_list, and element_list must match the order of the corresponding placeholders in prompt.
Examples
Text-to-Video
{
"model": "kling-v3-omni",
"prompt": "A golden retriever running on a beach at sunset, cinematic quality",
"mode": "std",
"duration": 5,
"aspect_ratio": "16:9"
}Image Reference
{
"model": "kling-v3-omni",
"prompt": "Make the person in <<<image_1>>> wave at the camera",
"mode": "pro",
"duration": 5,
"metadata": {
"image_list": [
{"image_url": "https://example.com/portrait.jpg"}
]
}
}First and End Frame
{
"model": "kling-v3-omni",
"prompt": "Transition from <<<image_1>>> to <<<image_2>>>",
"mode": "pro",
"duration": 5,
"metadata": {
"image_list": [
{"image_url": "https://example.com/start.jpg", "type": "first_frame"},
{"image_url": "https://example.com/end.jpg", "type": "end_frame"}
]
}
}Reference Video Input
{
"model": "kling-v3-omni",
"prompt": "Replace the background in <<<video_1>>> with a beach sunset",
"mode": "std",
"video_list": [
{
"video_url": "https://example.com/source-video.mp4",
"refer_type": "base",
"keep_original_sound": "no"
}
]
}Subject plus Motion Reference
{
"model": "kling-v3-omni",
"prompt": "Make <<<element_1>>> imitate the motion from <<<video_1>>>",
"mode": "pro",
"video_list": [
{
"video_url": "https://example.com/motion-reference.mp4",
"refer_type": "feature",
"keep_original_sound": "no"
}
],
"metadata": {
"element_list": [
{"url": "https://example.com/character.jpg", "type": "image", "role": "subject"}
]
}
}Video generation is asynchronous. Use the Get Video Task 상태 endpoint to query progress and results.
요청 예시
curl --request POST \
--url https://toapis.com/v1/videos/generations \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '{
"model": "kling-v3-omni",
"prompt": "Make the person in <<<image_1>>> wave at the camera",
"mode": "std",
"duration": 5,
"metadata": {
"image_list": [{"image_url": "https://example.com/portrait.jpg"}]
}
}'import requests
response = requests.post(
"https://toapis.com/v1/videos/generations",
headers={"Authorization": "Bearer <token>", "Content-Type": "application/json"},
json={
"model": "kling-v3-omni",
"prompt": "Make the person in <<<image_1>>> wave at the camera",
"mode": "std",
"duration": 5,
"metadata": {"image_list": [{"image_url": "https://example.com/portrait.jpg"}]},
},
)
print(response.json())const response = await fetch("https://toapis.com/v1/videos/generations", {
method: "POST",
headers: {
Authorization: "Bearer <token>",
"Content-Type": "application/json"
},
body: JSON.stringify({
model: "kling-v3-omni",
prompt: "Make the person in <<<image_1>>> wave at the camera",
mode: "std",
duration: 5,
metadata: {
image_list: [{ image_url: "https://example.com/portrait.jpg" }]
}
})
});
console.log(await response.json());