Generates robot manipulation actions from visual observations and text instructions using a 1.5B vision-language-action model. Uses streaming context and visual-token compression to cut per-step compute, runs a unified policy across tasks, and is open-sourced on Hugging Face under Apache-2.0.
Predicts eight future [x,y,yaw] waypoints for language-conditioned embodied person-following using fused DINOv3 and SigLIP visual features; trained with quality-driven, DAgger-style self-evolving data and optimized for on-device inference (~5+ FPS, ~180 ms).