Kling V3 Omni Image

Kling V3 Omni image model supporting single-image and series generation with flattened image references, element references, and up to 4K output.

POST /kling-v3-omni-image/async

cURL

single_result

curl --request POST \
  --url https://api.modellix.ai/api/v1/kling/kling-v3-omni-image/async \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "prompt": "Create a luxury perfume campaign visual using the reference bottle image"
}
'

200

{
  "code": 0,
  "message": "success",
  "data": {
    "status": "pending",
    "task_id": "task-v3-omni-image-123",
    "model_id": "model-123",
    "get_result": {
      "method": "GET",
      "url": "https://api.modellix.ai/api/v1/tasks/task-v3-omni-image-123"
    }
  }
}

Authorizations

Authorization · string · header · required

API Key authentication. Format: Bearer YOUR_API_KEY.

Body

application/json

Conditional rules: when result_type=series, omit n and provide series_amount; when result_type=single, omit series_amount and optionally use n. The combined count of images and element_ids must not exceed 10.

prompt · string · required

Image generation prompt, supports Chinese and English

Required string length: 1 - 2500

Example:

"Create a luxury perfume campaign visual using the reference bottle image"

images · string[]

Reference image list. The combined count of images and element_ids must not exceed 10.

Required array length: 1 - 10 elements

Reference image URL or Base64 encoded data

Minimum string length: 1

element_ids · integer[]

Flat element identifier list. The combined count of images and element_ids must not exceed 10.

Maximum array length: 10

Element identifier returned by Kling element workflows

resolution · enum<string> · default: 1k

Output resolution. Kling V3 Omni supports 1K, 2K, and 4K output.

Available options: 1k, 2k, 4k

Example:

"4k"

result_type · enum<string> · default: single

Result mode. Use single for one image or series for a multi-image sequence.

Available options: single, series

Example:

"series"

n · integer

Number of images to generate when result_type is single. This field is invalid when result_type is series.

Required range: 1 <= x <= 9

Example:

1

series_amount · integer

Number of frames/images to generate when result_type is series. This field is invalid when result_type is single.

Required range: 2 <= x <= 9

Example:

4

aspect_ratio · enum<string> · default: auto

Image aspect ratio. auto lets the model choose an intelligent aspect ratio.

Available options: 16:9, 9:16, 1:1, 4:3, 3:4, 3:2, 2:3, 21:9, auto

Example:

"auto"

Response

200

application/json

Task submitted successfully

code · integer · required

Response code, 0 indicates success

Example:

0

message · string · required

Response message

Example:

"success"

data · object · required

Detailed data about the submitted asynchronous task.

Show child attributes