🧠 Echo-Memory: Action-Conditioned World Model

Generate a video from an initial frame, a text prompt, and a camera rotation action. Based on the Echo-Memory paper — a controlled study of memory in action world models using the Wan 2.1 1.3B backbone.

Model: Echo-Team/Echo-Memory · Backbone: Wan-AI/Wan2.1-T2V-1.3B · Paper: arXiv:2606.09803 · Code: GitHub