static demo · no backend · drag to explore the point cloud
viewfinder · model input preview holes will be filled in by wan

image

camera drag to orbit · right-drag to pan · scroll to zoom

these come from your viewer camera. the green box on the left is what wan would receive as input.

generate

this is a static demo of the UI — generation is disabled. results below were produced by the live pipeline at x=0.15, z=0.1, 4 steps (~12 minutes on an RTX 4090).

[done] sample run · 4/4 steps · pre-rendered

result