curl --request POST \
--url https://api.poyo.ai/api/generate/submit \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '
{
"model": "wan3.0-text-to-video",
"callback_url": "https://your-domain.com/callback",
"input": {
"prompt": "A kitten runs across a moonlit rooftop while neon signs glow in the distance.",
"resolution": "720p",
"aspect_ratio": "16:9",
"duration": 5,
"audio": true
}
}
'{
"code": 200,
"data": {
"task_id": "task-unified-1787654321-abcd1234",
"status": "not_started",
"created_time": "2023-11-07T05:31:56Z"
}
}Wan
Wan 3.0 Video Generation
Wan 3.0 multimodal video generation API
POST
/
api
/
generate
/
submit
curl --request POST \
--url https://api.poyo.ai/api/generate/submit \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '
{
"model": "wan3.0-text-to-video",
"callback_url": "https://your-domain.com/callback",
"input": {
"prompt": "A kitten runs across a moonlit rooftop while neon signs glow in the distance.",
"resolution": "720p",
"aspect_ratio": "16:9",
"duration": 5,
"audio": true
}
}
'{
"code": 200,
"data": {
"task_id": "task-unified-1787654321-abcd1234",
"status": "not_started",
"created_time": "2023-11-07T05:31:56Z"
}
}- After submission, a
task_idis returned. Ifcallback_urlis provided, a POST request is sent when the task finishes or fails. - You can always retrieve the result through the unified Query Task Status endpoint.
Wan 3.0 Video Generation
Wan 3.0 includes three models for text, start/end frame, and multimodal reference video generation.Models
wan3.0-text-to-video— generates video from a required text promptwan3.0-image-to-video— generates video from one or two start/end frame imageswan3.0-reference-to-video— generates video from multimodal references
Key Features
- Duration: 2–30 seconds
- Resolution: 480p, 720p, or 1080p
- Start/End Frames: Use one or two
image_urlswith the image model - Generated Audio: Control the output audio track with
audio - Multimodal References: Use images, videos, audio clips, documents, or public webpages with the reference model
- Reference Labels: Mention inputs in the prompt as
Image1,Video1, orAudio1
Authorizations
Send your API key as Authorization: Bearer YOUR_API_KEY.
Body
application/json