Skip to main content
POST
/
v1
/
ai
/
reference-to-video
/
kling-v3-omni-std
Kling 3 Omni Standard - Video-to-video generation
curl --request POST \
  --url https://api.magnific.com/v1/ai/reference-to-video/kling-v3-omni-std \
  --header 'Content-Type: application/json' \
  --header 'x-magnific-api-key: <api-key>' \
  --data '
{
  "video_url": "<string>",
  "webhook_url": "<string>",
  "prompt": "<string>",
  "image_url": "<string>",
  "duration": "5",
  "aspect_ratio": "16:9",
  "cfg_scale": 0.5,
  "negative_prompt": "blur, distort, and low quality"
}
'
"<unknown>"

Authorizations

​
x-magnific-api-key
string
header
required

Your Magnific API key. Required for authentication. Learn how to obtain an API key

Body

application/json

Generate video using Kling 3 Omni with a reference video for motion/style guidance.

Required: The video_url parameter is required for this endpoint. Reference the video in your prompt using @Video1.

Best for:

  • Transferring motion patterns from reference videos
  • Maintaining visual consistency with reference material
  • Creating videos that follow a specific style or movement pattern
​
video_url
string<uri>
required

Required. URL of the reference video to use as a creative guide for video-to-video generation. Reference in your prompt as @Video1.

Video constraints:

  • Duration: 3-10 seconds
  • Resolution: 720-2160px (minimum 720px width or height)
  • Max file size: 200MB
  • Frame rate: 24-60 FPS
  • Formats: .mp4 or .mov only
​
webhook_url
string<uri>

Optional callback URL that receives asynchronous notifications when the task changes status. The payload includes the task status and result URL when completed.

​
prompt
string

Text prompt describing the desired video content. Maximum 2500 characters. Reference the video in your prompt as @Video1.

Tips for better results:

  • Be specific about actions, camera movements, and mood
  • Reference @Video1 to indicate how the reference video should influence generation
Maximum string length: 2500
​
image_url
string

URL of the start frame image for image-to-video generation with video reference.

Image requirements:

  • Minimum: 300x300 pixels
  • Maximum: 10MB file size
  • Formats: JPG, JPEG, PNG
​
duration
enum<string>
default:5

Duration of the generated video in seconds.

Range: 3-15 seconds Note: When using multi-shot mode, total duration across all shots cannot exceed 15 seconds.

Available options:
3,
4,
5,
6,
7,
8,
9,
10,
11,
12,
13,
14,
15
​
aspect_ratio
enum<string>
default:16:9

Aspect ratio for Kling V3 Omni video generation:

  • auto: Automatically match the input image aspect ratio (image-to-video only)
  • 16:9: Landscape (widescreen)
  • 9:16: Portrait (vertical)
  • 1:1: Square
Available options:
auto,
16:9,
9:16,
1:1
​
cfg_scale
number<float>
default:0.5

Guidance scale controlling prompt adherence versus creative freedom. Range: 0 to 1. Values above 1 are not supported for Kling 3 (only Kling 1.x models support cfg_scale up to 2).

  • 0: Maximum creative freedom, loosest interpretation of the prompt
  • 0.5 (default): Balanced between prompt adherence and creativity
  • 1: Strongest adherence to the prompt, least creative variation
Required range: 0 <= x <= 1
​
negative_prompt
string
default:blur, distort, and low quality

Undesired elements to avoid in the generated video. Maximum 2500 characters.

Maximum string length: 2500

Response

OK - Task created successfully

OK - The task exists and the status is returned