🔥 We release RynnVLA-001-7B-Trajectory (Stage 2: Human-Centric Trajectory-Aware Video Modeling), which is further trained to predict future human trajectories and frames. This stage is initialized from
RynnVLA-001-7B-Base.
RynnVLA-001 is a VLA model based on pretrained video generation model. The key insight is to implicitly transfer manipulation skills learned from human demonstrations in ego-centric videos to the manipulation of robot arms.
