Expressive character performance
Lifelike facial expressions, body movement and emotional reactions — from quiet conversations to intense action.
Animate a single image or combine up to 15 reference images with voice references for consistent characters — synchronized audio, expressive performance and up to 16-second 4K clips.
Vidu’s flagship video model built for cinematic visuals — expressive characters, realistic motion, native audio and reference-driven consistency from images and voices.
From a single still to a fully referenced scene, Vidu Q4 keeps performance, identity and sound under your direction.
Lifelike facial expressions, body movement and emotional reactions — from quiet conversations to intense action.
Up to 15 reference images control character appearance, clothing, objects and scene style across new angles and actions.
Attach up to 3 audio clips to guide character voices, tone and emotional delivery with synchronized lip movement.
Choose 540p for quick drafts up to 4K for final renders — 10-bit color depth and rich detail at every tier.
Tracking shots, dramatic close-ups, chases and smooth scene transitions while preserving visual continuity.
Dialogue, ambient sound and cinematic effects generated with the video — or switch it off for silent clips.
Pick a workflow, attach your materials and generate.
Image to Video animates one still; Reference to Video combines up to 15 images and 3 voice clips for consistency.
Add your image or references, then describe the scene — referring to materials by order like “reference image 2”.
Set aspect ratio (reference mode), resolution from 540p to 4K, duration up to 16s and the audio switch.
The studio estimates your cost live. Preview the result inline and download the finished MP4.
Answers to common questions about inputs, settings and pricing.
Vidu Q4 is Vidu’s flagship cinematic video model. It generates expressive character performance with synchronized native audio, up to 4K resolution and 16-second clips.
Image to Video animates a single image as the visual foundation of the shot. Reference to Video accepts 1 to 15 reference images plus up to 3 voice references so characters, clothing, objects and scene styles stay consistent across new actions and camera angles.
Refer to inputs by order in natural language, for example “the man in reference image 2 says … using reference audio 1”. Reference images and audios keep their upload order.
Images: JPG, JPEG, PNG or WEBP, up to 50MB each — one image in Image to Video mode, 1–15 in Reference mode. Voice references: up to 3 MP3 files, each 3–12 seconds and up to 50MB.
Choose 540p, 720p, 1080p, 2K or 4K output, with clip lengths from 3 to 16 seconds. Image to Video inherits the aspect ratio of the input image; Reference mode offers 16:9, 9:16, 1:1, 4:3 and 3:4.
Cost is a per-second rate multiplied by the output duration: 10 gems/s at 540p, 20 at 720p, 30 at 1080p, 60 at 2K and 120 at 4K. The audio switch does not change the price and reference images or voice inputs are billed at no extra charge.
Yes by default. Vidu Q4 generates synchronized dialogue and sound effects; turn off the Audio toggle for a silent clip at the same price.
Start with a single image or a full reference set and generate cinematic 4K video with Vidu Q4.
Start creating with Vidu Q4