VLA Models

VLA (Vision-Language-Action) – “VLM + action output": see and reason plus act with hands. Two-level architecture: slow long-term planning (what to do – breaking down the instruction into subgoals) and fast VLA control (how to do – continuous control signals). Two directions of action representation: discrete tokens (RT-2, OpenVLA – autoregression as text generation) and continuous trajectory (π₀ – flow matching, smoother for precise manipulations). The foundation is imitation learning on cross-platform demonstrations; RL – an additional layer. Action decomposition into blocks: one model output – a sequence of future actions, execution at high frequency.

Related: Computer Use, Sim2Real, [Two-level architecture planning-control]