Edge Dance (hosted at edge‑dance.github.io) is an AI research‑based tool that automatically creates dance motion sequences from music input. Powered by a transformer‑based diffusion model paired with Jukebox — a sophisticated music embedding extractor — the platform interprets rhythms and beats to produce high‑quality dance animations that stay synchronized with the chosen track. The system supports editable choreography, letting users define constraints like joint‑wise body motion or specific movement segments, and it uses an innovative Contact Consistency Loss to keep body movements physically realistic (like avoiding unnatural foot sliding). Though originally developed as part of academic research at Stanford University and showcased under a CVPR 2023 project, Edge Dance can be used by creators, animators, game designers, choreographers, and digital artists to generate expressive dance sequences from any piece of music.