AI Models
What each model is good at, where it differs from the others, and what it costs to run.
Alibaba's Wan 3.0 video model. A longer clip from a prompt or two frames, with optional sound, or from reference pictures you name.
Credits come with a Generator plan. The exact cost is shown on the Generate button before anything is spent.
Wan 3.0 is Alibaba's later Wan video model. Use it when a Wan 2.7 clip is not enough room for the shot, or when the clip should carry sound.
Alibaba describes Wan 3.0 as longer clips from a richer set of inputs, with steadier faces, motion, and scenes, and with voice generated along with the picture. On Visxy the inputs are the prompt, frames, sound, and reference pictures.
Write the shot. Turn sound on if you want it. Give one picture as the opening frame and an optional second as the closing frame. Or open Wan 3.0 Reference, attach two to ten pictures, and name them @image1 onward in the prompt.
Wan 2.7 has Enhance and no sound switch, and its reference mode takes fewer pictures. Wan 3.0 Prime is the speed-oriented sibling with the same controls. HappyHorse 1.1 is Alibaba's other video model, and it starts from one opening picture rather than two frames.
Wan 3.0 Reference takes two to ten pictures. Name them @image1 onward in the prompt.
Yes. The switch starts off until you turn it on.
Alibaba describes Prime as the speed-optimized Wan 3.0, aimed at keeping quality while generating faster. The controls match.
No. Enhance is on Wan 2.7, not on Wan 3.0.
