Luma Ray 2

Overview

Luma Ray 2 is a high-quality video generation model that supports both text-to-video and image-to-video modes.


Test API KEY: key_00000000000000000000000000000000

Endpoints

1. Create Task

Request


POST https://api.pixwith.ai/api/task/create

Headers

Header	Value	Required
Content-Type	application/json	Yes
Api-Key	{API_KEY}	Yes

Request Parameters

Parameter	Type	Required	Description
prompt	string	Yes	The input prompt (e.g., “A cat walking in the garden”)
image_urls	array	Conditional	[“https://cdn.pixwith.ai/xxx.jpg ”]. Provide 1 image (start_image) or 2 images (start_image and end_image). Second image is optional.
model_id	string	Yes	`2-4`
options	object	Yes	Configuration options for the task
options.prompt_optimization	boolean	Yes	If true, all prompt words will be automatically translated into English.
options.aspect_ratio	string	Yes	Aspect ratio of the output. Valid values: `'16:9'`, `'9:16'`, `'1:1'`, `'4:3'`, `'3:4'`, `'9:21'`, or `'21:9'`
options.resolution	string	Yes	Resolution of the output. Valid values: `'540p'` or `'720p'`
options.duration	int	Yes	Duration in seconds. Valid values: `5` or `9`
options.loop	boolean	No	Loop the video. Default: `false`

Example Request - Text-to-Video


{
  "prompt": "A cat walking in the garden",
  "model_id": "2-4",
  "options": {
    "prompt_optimization": true,
    "aspect_ratio": "16:9",
    "resolution": "540p",
    "duration": 5,
    "loop": false
  }
}

Example Request - Image-to-Video (with start and end images)


{
  "prompt": "A cat walking in the garden",
  "image_urls": [
    "https://cdn.pixwith.ai/images/cf0ddb5ad6a0f0c85ab4d1dceb9fd89c_0.jpeg",
    "https://cdn.pixwith.ai/images/cf0ddb5ad6a0f0c85ab4d1dceb9fd89c_1.jpeg"
  ],
  "model_id": "3-4",
  "options": {
    "prompt_optimization": true,
    "aspect_ratio": "16:9",
    "resolution": "540p",
    "duration": 5,
    "loop": false
  }
}

Response Parameters

Parameter	Type	Description
code	int	1=success, 0=error
message	string	Success and failure messages
data	object
data.task_id	string
data.estimated_time	int	Estimated time in seconds

Response


{
  "code": 1,
  "message": "success",
  "data": {
    "task_id": "d425ab4dc256f0ebaa84862b9cea8a3d",
    "estimated_time": 42
  }
}