Vidu Q3 (Reference)
Vidu Q3 reference-image to video — the standard quality tier. Supply 1-7 reference images via image_urls; the model keeps the subjects consistent across the generated video.
This model only accepts image_urls. For text-to-video, image-to-video or start-end frame generation use viduq3_pro — passing image_url or image_end_url here returns an error.
Why not
viduq3_prowith reference images? Vidu’s reference-to-video endpoint does not accept theviduq3-promodel at all. This model is the standard-tier equivalent on that endpoint;viduq3_turbo_referenceis the faster, cheaper tier.
Pricing — per second of generated video, in tokens (USD at $0.02/token):
| resolution | tokens/s | USD/s |
|---|---|---|
| 540p | 1.75 | $0.035 |
| 720p | 3 | $0.06 |
| 1080p | 3.75 | $0.075 |
Example: an 8s 720p video costs 24 tokens.
Returns an img_uuid; poll GET /api/v1/jobs/detail (or use custom_callback_url) to get the generated video.
Headers
Body
Model name. Must be viduq3_reference.
viduq3_reference Text prompt describing the video.
Reference image URLs (required, 1-7 images). Each image must be at least 128x128 px, aspect ratio between 1:4 and 4:1, max 50MB; png/jpeg/jpg/webp.
1 - 7 elementsVideo duration in seconds (3-16). Defaults to 5.
3 <= x <= 16Output resolution.
540p, 720p, 1080p Aspect ratio of the output video. Reference-to-video supports fewer ratios than the generation endpoints. aspect_ratio is accepted as a compatible alias.
16:9, 9:16, 1:1 Optional. A publicly accessible HTTPS URL. When the task status changes (processing / success / failed), GoEnhance sends a POST request to this URL. The request body is identical to the response of GET /api/v1/jobs/detail. If your server does not respond with HTTP 200, the notification is retried up to 3 times, with a 3-second timeout per attempt.
"https://your-server.com/goenhance/callback"
