Vidu AI

Vidu is an AI video generator that creates clips from a text prompt, from a still image, or from reference material that fixes a subject's appearance across shots. Reference to video is the distinguishing mode: supply images of a character, a costume or a prop and the model holds those details consistent through cuts and camera moves, which is the part most text to video tools handle poorly. Text to image and reference to image modes feed the same pipeline, so a look can be designed as a still before it is animated. The current flagship model is Vidu Q4, with Q3 still selectable, and a separate real time model generates at conversational speed for interactive use. Vidu Agent takes a one sentence brief and works it up into a finished piece, handling the intermediate steps itself. Around the models sit a template library for common social formats and effects, a broader creative toolkit, and mobile apps. The audience runs from solo creators making short form social content and advertising through to animation teams, who use the model to fill the gaps between keyframes and references. An API platform exposes the same models to developers, and enterprise plans add team collaboration on top of the individual tiers. Vidu publishes individual and enterprise subscription plans with monthly and annual billing, plus pay as you use API access billed on consumption. The amounts are not readable as published text: the pricing page loads its figures dynamically and the page source carries no plan prices or structured pricing data, so current rates can only be seen in a browser with scripts running.