Drop files here, paste files, or add a URL. Disabled while reference images or audio are set. Remove them to use a start frame.
Drop files here, paste files, or add a URL. 1–12 images of characters, products, objects or locations, or use a start frame instead. PNG, JPEG or WebP, up to 50 MB each.
Drop files here, paste files, or add a URL. Optional. Requires at least one reference image. Up to 3 MP3 voice clips, 3–12 seconds and up to 50 MB each, to keep character voices consistent.
Example Output
This is sample data. Generate a new result to see real output.
Transparent pricing with no hidden fees. Pay as you go.
vidu-q4vidu-q4vidu-q4vidu-q4vidu-q4| Model | Spec | PoYo price | Official price | You save |
|---|---|---|---|---|
| vidu-q4 | 540p | $0.045/sec 9 credits /sec | - | - |
720p | $0.095/sec 19 credits /sec | - | - | |
1080p | $0.120/sec 24 credits /sec | - | - | |
2K | $0.190/sec 38 credits /sec | - | - | |
4K | $0.390/sec 78 credits /sec | - | - |
* Actual fees are based on the final output.
Start frames, reference images and voice clips into cinematic video with sound
Send exactly one image in image_urls and Vidu Q4 animates it as the first frame. The prompt is optional and can describe motion, camera movement and sound. In this mode, the output follows the image's aspect ratio, so aspect_ratio and generate_audio are not accepted.
Send 1–12 images in reference_image_urls and up to 3 voice clips in reference_audio_urls, then describe the scene in the required prompt. Refer to each input as [@reference_image_1] or [@reference_audio_1], choose the aspect ratio and turn native audio generation on or off.
Official Vidu Q4 samples show cinematic detail, consistent characters and large-scale action with synchronized sound.
01
Vidu Q4 generates picture and sound together, including dialogue, music, ambience and sound effects matched to on-screen action. Close-ups keep fine textures such as skin, hair, fabric and instrument strings, so a single request can produce a clip that already feels finished.
02
Upload character, outfit, prop or location images and reference them directly in your prompt. Vidu Q4 keeps faces, costumes and visual style stable across camera angles and shots, which makes it practical for recurring characters in series, ads and branded content.
03
Vidu Q4 is built for fast motion, martial arts, crowds and visual effects without losing physical coherence. Describe several beats in one prompt and the model can arrange them as a multi-shot sequence of up to 16 seconds, with camera changes that follow the story.
Keep the same cast across episodes by reusing character and location references. Combine reference images with voice clips to produce dialogue scenes, then assemble 16-second multi-shot clips into longer stories.
Turn product photos, packaging and brand characters into cinematic spots with music and sound effects. Choose 16:9, 9:16, 4:3, 3:4 or 1:1 in reference mode to match each placement.
Animate a single key image into a short clip with sound, or place your recurring persona into new scenes. Lower resolutions keep drafts affordable before you render the final version in 1080p, 2K or 4K.
Prototype action sequences, camera moves and VFX moments from concept art. Use the start-frame mode to test how a specific frame evolves, or reference mode to block out whole scenes.
Bring character sheets and environment concepts to life for trailers, cutscene drafts and pitch videos while keeping designs consistent across shots.
Add Vidu Q4 to your product through PoYo's asynchronous API. Submit a task, poll its status or receive a callback, and charge your users per second of output.
Vidu Q4 raises the resolution ceiling to 4K and accepts more reference images than Vidu Q3, while keeping native audio and multi-shot generation. Clips still run up to 16 seconds.
| Feature | Vidu Q4 on PoYo | Vidu Q3 |
|---|---|---|
| Generation modes | Image to video and reference to video | Text to video, image to video, start-end frames and reference to video |
| Duration | 3–16 seconds | Up to 16 seconds |
| Resolution | 540p, 720p, 1080p, 2K or 4K | 540p, 720p or 1080p |
| Reference images | 1–12 images | 1–7 images |
| Voice references | Up to 3 audio clips | Up to 3 MP3 clips of 3–12 seconds (viduq3-mix) |
| Native audio | Yes, optional in reference to video | Yes, on by default |
| Multi-shot | Supported | Supported through intelligent camera switching |
The Vidu Q4 column lists the parameters PoYo accepts. Text-only generation is not offered for Vidu Q4 on PoYo, and ShengShu's own API allows up to 15 reference images. The Vidu Q3 column is based on the official Vidu API documentation for the Q3 model family.
Step 1: Register and create an API Key.
Sign in to PoYo and create an access key for your application.
Get API Key →
Step 2: Top up credits.
Add credits to your account. The Playground shows the estimated credits for your chosen resolution and duration before you generate.
Add Credits →
Step 3: Integrate the API.
Send model: "vidu-q4" with either image_urls or reference_image_urls to /api/generate/submit, then query the task ID or provide a callback URL for completion notifications.
View API Reference →
Choose the resolution and duration that fit each project. The pricing table and the Playground estimate use the same per-second rates, with no subscription required.
Upload a start frame or reference images and voice clips, switch between form and JSON editing, and check the exact request before integrating.
Use the same authentication, submission and status workflow for Vidu Q4 and other video, image and audio models on PoYo.
Poll the task ID for progress or provide a callback URL, then retrieve the finished video URL through the same asynchronous workflow.
