AIREITER
Vidu Q4 AI Video Generator and API
Vidu Q4 is a video model from Vidu (ShengShu Technology). It animates one first-frame image, or combines up to 15 reference images and 3 reference audio clips, into a 3 to 16 second video with synchronized dialogue and sound effects.
Two input modes
First Frame animates a single image. Reference builds a new shot from 1–15 reference images and optional reference audio.
3 to 16 seconds
Pick any whole-second length from 3 to 16. The default is 5 seconds.
540p up to 4K
Render at 540p, 720p, 1080p, 2K, or 4K. The default is 720p.
Synced audio by default
Dialogue and sound effects are generated with the video. Turn audio off for a silent clip.
First Frame vs Reference Mode
Vidu Q4 always starts from at least one image. Choose the mode by what you already have.
First Frame
Upload exactly one image as the opening frame. The prompt is optional, and the video keeps the image's aspect ratio.
Reference
Upload 1–15 reference images and write a prompt (required). Choose 16:9, 9:16, 1:1, 4:3, or 3:4.
Reference audio
Reference mode only. Add up to 3 MP3 clips, each 3–12 seconds, to guide voice and tone.
What Vidu Q4 does not do
There is no text-only mode and no first-and-last-frame mode. Every request needs at least one image.
What to Create with Vidu Q4
Bring a still image to life
Turn a product shot, illustration, or keyframe into a moving clip that starts exactly on your image.
Compose a scene from several references
Give the character, the outfit, and the location as separate reference images, then describe how they come together in one shot.
Voiced clips with a reference voice
Add a short MP3 so the generated dialogue follows the voice and tone of your reference audio.
Vertical, square, and widescreen
In Reference mode, generate in 9:16, 1:1, 16:9, 4:3, or 3:4 so the clip fits its destination without cropping.
Supported Vidu Q4 Controls
The playground and the API expose the same parameters.
Prompt
Describe the motion and the scene. Optional in First Frame, required in Reference. Refer to inputs by order, for example the man in reference image 2, using reference audio 1.
Generation type
frame (First Frame, the default) or reference (Reference).
Images
First Frame: exactly 1 image. Reference: 1–15 images. JPG, PNG, or WEBP, up to 50MB each.
Reference audio
Reference mode only. 0–3 MP3 clips, 3–12 seconds each, up to 50MB each.
Resolution and length
540p, 720p (default), 1080p, 2K, or 4K. Length is 3–16 whole seconds, default 5.
Aspect ratio and audio
Aspect ratio applies to Reference mode only: 16:9 (default), 9:16, 1:1, 4:3, or 3:4. Generate Audio is on by default and can be turned off.
How to Generate a Vidu Q4 Video
Four steps, in the browser or through the API.
Choose a mode
First Frame to animate one image, or Reference to build a shot from several images and optional audio.
Upload your inputs
One image for First Frame. For Reference, 1–15 images and up to 3 MP3 clips of 3–12 seconds.
Write the prompt
Describe motion, camera, and dialogue. In Reference mode, name inputs by order, such as reference image 2 or reference audio 1.
Set specs and generate
Pick resolution, length, and, in Reference mode, aspect ratio. Keep audio on for dialogue and sound effects, then generate.
Use Vidu Q4 through the AIReiter API
Submit one task and query the same external task ID until it reaches a terminal state.
Model ID
Send vidu_q4 as the model value in the request body.
Submit a task
POST /api/openapi/submit with the model, a unique out_task_id, and params: type (frame or reference), prompt, image_url, audio_url, resolution, video_length, aspect_ratio, and generate_audio.
Query the result
POST /api/openapi/query with the same out_task_id until the status is completed or failed. Completed responses include the video URL.
Input rules
type frame takes exactly 1 image and no audio. type reference takes 1–15 images, 0–3 MP3 clips, and a required prompt. aspect_ratio applies to reference only.
Related AI Video Models on AIReiter
Kling v3 Omni
Text, multi-image reference, first/last frames, and reference video up to 15s.
Wan 3.0
Text, first/last frames, mixed references, and document-to-video up to 30 seconds.
Gemini Omni 1.1 Flash Ext
4, 6, 8, or 10 second clips at 720p, 1080p, or 4K from text, up to 7 stills, or a reference clip.
AI Video Generator
Switch between every video model on AIReiter from one tool.
Vidu Q4 FAQ
Answers about modes, length, resolution, reference audio, and API usage.
/ 01What is Vidu Q4?
Vidu Q4 is a video model from Vidu (ShengShu Technology). On AIReiter it animates one first-frame image, or combines 1–15 reference images with up to 3 reference audio clips, into a 3–16 second video at 540p to 4K with synchronized audio.
/ 02Does Vidu Q4 support text-to-video?
No. Every request needs at least one image: one image in First Frame mode, or 1–15 reference images in Reference mode.
/ 03Does Vidu Q4 support first and last frames?
No. First Frame mode takes a single opening image. There is no end-frame input.
/ 04How long can a Vidu Q4 video be?
From 3 to 16 seconds in whole seconds. The default is 5 seconds.
/ 05What resolutions does Vidu Q4 support?
540p, 720p, 1080p, 2K, and 4K. The default is 720p.
/ 06How do I use reference audio?
Switch to Reference mode and upload up to 3 MP3 clips, each 3–12 seconds. Mention them in the prompt by order, such as using reference audio 1, to guide the voice and tone.
/ 07How do I point to a specific reference image?
Refer to images by upload order in the prompt, for example the man in reference image 2 walks toward the car in reference image 1.
/ 08Which aspect ratios are supported?
Reference mode supports 16:9, 9:16, 1:1, 4:3, and 3:4. First Frame mode keeps the aspect ratio of your image.
/ 09Which API model ID should I use?
Use vidu_q4 with POST /api/openapi/submit, then poll POST /api/openapi/query with the same out_task_id.