curl --request POST \
--url https://api.poyo.ai/api/generate/submit \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '
{
"model": "vidu-q4",
"input": {
"image_urls": [
"https://example.com/start-frame.png"
]
}
}
'{
"code": 200,
"data": {
"task_id": "task-unified-1792190000-example",
"status": "not_started",
"created_time": "2026-10-17T10:30:00"
}
}{
"code": 400,
"error": {
"message": "Invalid request parameters",
"type": "invalid_request_error"
}
}{
"code": 400,
"error": {
"message": "Invalid request parameters",
"type": "invalid_request_error"
}
}Vidu
Vidu Q4
Generate a 3–16 second video from one starting frame or reference images, with optional reference audio, at resolutions from 540p to 4k.
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": "vidu-q4",
"input": {
"image_urls": [
"https://example.com/start-frame.png"
]
}
}
'{
"code": 200,
"data": {
"task_id": "task-unified-1792190000-example",
"status": "not_started",
"created_time": "2026-10-17T10:30:00"
}
}{
"code": 400,
"error": {
"message": "Invalid request parameters",
"type": "invalid_request_error"
}
}{
"code": 400,
"error": {
"message": "Invalid request parameters",
"type": "invalid_request_error"
}
}- After submission, a
task_idwill be returned. If you provided acallback_url, when the task status becomesfinishedorfailed, a POST request will be sent to thecallback_url. - Regardless of whether
callback_urlis provided, you can retrieve the response result through the unified Query Task Status endpoint.
Models
- vidu-q4 — Image-to-video and reference-to-video generation
Features
- Generate videos from 3 to 16 seconds long.
- Animate one starting-frame image or guide a video with 1–12 reference images and up to 3 optional reference audio files.
- Choose an output resolution from 540p to 4k.
- Generate an optional audio track in reference-to-video mode.
Input modes
- Image-to-video: Provide exactly one URL in
image_urls.promptis optional. Do not combine this mode withreference_image_urlsorreference_audio_urls, and do not sendaspect_ratioorgenerate_audio. - Reference-to-video: Provide
promptand 1–12 URLs inreference_image_urls.reference_audio_urlsis optional and must accompany reference images. Do not sendimage_urls.
Required parameters
prompt(string): Required for reference-to-video. Must contain 1–5,000 characters and cannot be blank.image_urls(array of strings): Required for image-to-video. Must contain exactly one starting-frame image URL.reference_image_urls(array of strings): Required for reference-to-video. Must contain 1–12 reference image URLs.
Optional parameters
prompt(string): Optional for image-to-video, with a maximum of 5,000 characters.reference_audio_urls(array of strings): Optional audio accompanyingreference_image_urls, up to 3 files. Each file must be MP3 and 3–12 seconds long. Audio cannot be used without reference images.duration(integer): Output duration from 3 to 16 seconds. Defaults to5.resolution(string):540p,720p,1080p,2k, or4k. Defaults to720p.aspect_ratio(string):16:9,9:16,4:3,3:4, or1:1. Defaults to16:9. Available for reference-to-video only.generate_audio(boolean): Generate an audio track. Defaults tofalse. Available for reference-to-video only.enable_safety_checker(boolean): Enable content safety checks. Defaults totrue.
Examples
The request examples show animation of one starting frame with an optional prompt, and generation guided by reference images with optional audio. Both modes use the samevidu-q4 model ID.
Notes
- Put all generation parameters inside
input. - Images must use publicly accessible HTTP or HTTPS URLs.
- Reference audio must use publicly accessible HTTP or HTTPS URLs.
- Image-to-video rejects
aspect_ratioandgenerate_audioeven whengenerate_audioisfalse.
Authorizations
Bearer authentication header of the form Bearer <token>, where <token> is your auth token.
Body
application/json
Public model identifier.
Available options:
vidu-q4 Provide one image_urls starting frame, or a prompt with 1-12 reference_image_urls. Reference audio is optional and requires reference images.
Show child attributes
Show child attributes
Optional webhook URL for task completion or failure notifications.