Wan is Alibaba's line of video models. Wan 3.0 is the newest generation, capable of more than just animating a still image: it builds a mini story within the clip with shot changes, voices characters and the scene, and keeps the details of the source frame intact.
The model works in both modes. From a photo, Wan 3.0 creates a video while preserving the character's appearance, setting and fine details. From text, it shoots a scene from scratch in the format you need: 16:9, 9:16 or 1:1. Sound is always included: speech, ambient noise and music are generated together with the picture. Duration is 5 or 10 seconds, resolution is 720p or 1080p.
Wan 3.0 takes longer to render than other models, about 3 minutes, but it's great for short stories, dialogue-driven ads and clips that need shot changes. For fast, low-cost videos, choose MiniMax H3 Turbo.