generate-video
Create videos from text prompts and images using AI models, with options to control aspect ratio, resolution, duration, and looping for customized outputs.
Instructions
Generate a video from text prompt and/or images using AI models (Luma or Kling)
Input Schema
Name | Required | Description | Default |
---|---|---|---|
aspect_ratio | No | Aspect ratio of the video | 16:9 |
duration | No | Duration of the video (9s costs 2x more) | 5s |
end_image_url | No | Final image to end the video with (URL or base64 data URI) | |
image_url | No | Initial image to start the video from (URL or base64 data URI) | |
loop | No | Whether the video should loop (blend end with beginning) | |
model | No | AI model to use (luma=Ray2, kling=Kling) | luma |
prompt | Yes | Text description of the desired video content | |
resolution | No | Resolution of the video (higher resolutions use more credits) | 540p |
Input Schema (JSON Schema)
{
"properties": {
"aspect_ratio": {
"default": "16:9",
"description": "Aspect ratio of the video",
"enum": [
"16:9",
"9:16",
"4:3",
"3:4",
"21:9",
"9:21"
],
"type": "string"
},
"duration": {
"default": "5s",
"description": "Duration of the video (9s costs 2x more)",
"enum": [
"5s",
"9s"
],
"type": "string"
},
"end_image_url": {
"description": "Final image to end the video with (URL or base64 data URI)",
"type": "string"
},
"image_url": {
"description": "Initial image to start the video from (URL or base64 data URI)",
"type": "string"
},
"loop": {
"default": false,
"description": "Whether the video should loop (blend end with beginning)",
"type": "boolean"
},
"model": {
"default": "luma",
"description": "AI model to use (luma=Ray2, kling=Kling)",
"enum": [
"luma",
"kling"
],
"type": "string"
},
"prompt": {
"description": "Text description of the desired video content",
"type": "string"
},
"resolution": {
"default": "540p",
"description": "Resolution of the video (higher resolutions use more credits)",
"enum": [
"540p",
"720p",
"1080p"
],
"type": "string"
}
},
"required": [
"prompt"
],
"type": "object"
}