Sample Request

curl --request POST \
     --url https://cloud.leonardo.ai/api/rest/v2/generations \
     --header 'accept: application/json' \
     --header 'authorization: Bearer <YOUR_API_KEY>' \
     --header 'content-type: application/json' \
     --data '
{
    "model": "seedance-2.0",
    "public": false,
    "parameters": {
      "prompt": "A koala plays with a cat in a sunny garden",
      "guidances": {
        "start_frame": [
          {
            "image": {
              "id": "<YOUR_START_IMAGE_ID>",
              "type": "UPLOADED"
            }
          }
        ],
        "end_frame": [
          {
            "image": {
              "id": "<YOUR_END_IMAGE_ID>",
              "type": "GENERATED"
            }
          }
        ]
      },
      "duration": 8,
      "mode": "RESOLUTION_720",
      "prompt_enhance": "OFF",
      "width": 1280,
      "height": 720
    }
}
'

Recipe

API Request Endpoint, Headers, Parameters

Endpoint

https://cloud.leonardo.ai/api/rest/v2/generations

Headers

--header "accept: application/json" \
--header "authorization: Bearer <YOUR_API_KEY>" \
--header "content-type: application/json"

Body Parameters

Both Seedance 2.0 and Seedance 2.0 Fast models share the same parameters.

Parameter	Type	Definition
duration	`integer`	Length of the generated video in seconds. Set to `4`, `5`, `6`, `7`, `8`, `9`, `10`, `11`, `12`, `13`, `14`, or `15`. Defaults to `8`.
guidances.end_frame	`array`	Maximum 1 item. Each item includes image.id and image.type (`GENERATED` or `UPLOADED`). Only available when `start_frame` is provided.
guidances.image_reference	`array`	Maximum 4 items. Reference images to guide style and composition. Incompatible with `start_frame` and `end_frame`.
guidances.start_frame	`array`	Maximum 1 item. Each item includes image.id and image.type (`GENERATED` or `UPLOADED`).
guidances.video_reference_base	`array`	Maximum 3 items. Reference videos to guide motion. Incompatible with `start_frame` and `end_frame`.
height	`integer`	Sets the height of the output video in pixels. Set to `0` to automatically use the source image's aspect ratio (image-to-video only).
model	`string`	Specifies the model used for generation. Set to `seedance-2.0` or `seedance-2.0-fast`.
mode	`string`	Resolution tier. Set to `RESOLUTION_480` for Standard (480p) or `RESOLUTION_720` for HD (720p). Defaults to `RESOLUTION_720`.
motion_has_audio	`boolean`	When set to `true`, generates native audio including dialogue, narration, sound effects, and ambient sounds. Defaults to `false`.
prompt	`string`	Text description of the video content to generate. Maximum of 1500 characters.
prompt_enhance	`string`	Enables or disables prompt enhancement. Automatically disabled when `start_frame` is set.
public	`boolean`	Controls whether the generated video is public (`true`) or private (`false`).
quantity	`integer`	Number of videos to generate. Maximum 1. Defaults to `1`.
seed	`integer`	Random seed for reproducibility. Range: `0`–`4294967295`. Set to `-1` for random. Defaults to `-1`.
width	`integer`	Sets the width of the output video in pixels. Set to `0` to automatically use the source image's aspect ratio (image-to-video only).

Constraints

Both models share the same constraints.

end_frame requires start_frame to also be set.
guidances.image_reference is incompatible with start_frame and end_frame.
guidances.video_reference_base is incompatible with start_frame and end_frame.
Setting start_frame automatically disables prompt_enhance.

Dimensions

The API accepts width and height parameters in the following aspect ratio combinations. Both models use the same dimensions, determined by the mode parameter.

For image-to-video, set width: 0 and height: 0 to automatically use the source image's aspect ratio.

Resolution	Aspect Ratio	(Width x Height)
480p	21:9	992x432
	16:9	864x496
	4:3	752x560
	1:1	640x640
	3:4	560x752
	9:16	496x864
	9:21	432x992
720p (default)	21:9	1470x630
	16:9	1280x720
	4:3	1112x834
	1:1	960x960
	3:4	834x1112
	9:16	720x1280
	9:21	630x1470