Vidu Q4 Reference to Video builds a clip of 3 to 16 seconds from one to fifteen reference pictures, with optional reference audio and a shape selector.
This is the multi-image product of the Vidu Q4 family: instead of animating one picture, you supply a set of references and describe how they belong together. The single-image workflow is covered in the Vidu Q4 Image to Video guide.
Two separate audio controls exist here. The sound switch turns generated audio on or off, and the reference audio picker accepts up to three MP3 clips alongside your pictures. One does not replace the other.
Upload your references, add reference audio if you have it, then set length, quality tier, shape and the sound switch. A description is required in this mode, because the model needs to know how the references belong together.
Supply a single picture or a whole set and describe the scene they should produce together.
Add up to three MP3 clips of 3 to 12 seconds each as an audio hint next to the pictures, with the workspace checking every clip's duration.
Choose 16:9, 9:16, 1:1, 4:3 or 3:4, which makes this the Vidu Q4 product to use when the destination needs a shape your image does not have.
Supply several pictures of the same subject, describe a new scene for it and compare the results before paying for higher tiers.
Combine product and location references, then describe their arrangement in the scene. Check product details in the generated result.
Attach a voice or ambience clip and describe how it should relate to the scene. The result may not reproduce the supplied audio exactly.
Use a button on this page. It opens the right workspace with this model already selected, in the language you are reading now.
Write what should happen in the shot or image. Concrete subjects, framing and action give the model more to work with than short labels.
Choose length or quality tier, shape and sound where offered, then read the credit quote for those exact settings before you generate.
Generation uses account credits. Cost depends on the model, quality and output size or duration. Reference images, video or audio may add charges in some modes. Check the workspace quote for your selected inputs and settings before generating; the pricing page lists credit packages.
See credit packagesBetween one and fifteen per task, in JPEG, PNG or WebP and up to 50 MB each.
Up to three MP3 clips of 3 to 12 seconds each, up to 50 MB per file, plus a sound switch for generated audio. The two inputs are separate.
Yes. Reference mode needs a description of how the images belong together, up to 20000 characters.
16:9, 9:16, 1:1, 4:3 and 3:4. The single-image Vidu Q4 product has no shape selector; this one does.
From 3 to 16 seconds in whole seconds, at 540p, 720p, 1080p, 2k or 4k.
Other models on this platform that do a similar job, or that pair well with this one.
Vidu Q4 animates exactly one starting image into a clip of 3 to 16 seconds, in five quality tiers from 540p to 4k, with optional sound.
Read the guideKling 4.0 Reference builds a clip of 3 to 30 seconds from named materials: reference images, reference clips, subjects and voices, each addressed by name in the prompt.
Read the guideMiniMax H3 turns a description or a single starting image into a clip of 4 to 15 seconds that carries a generated sound track.
Read the guide