From language to physical intelligence

This sequence is a reading path, not a mandatory architecture; systems may skip layers or create feedback loops.

1. 01Large Language Model

   Models statistical structure in language and produces or interprets text.

2. 02Vision-Language Model

   Connects visual observations with language representations and instructions.

3. 03World Model

   Represents how an environment may change, including the consequences of actions.

4. 04Planner

   Searches possible actions or trajectories against a goal and predicted outcomes.

5. 05Agent

   Observes, chooses, acts, and updates its behavior across a feedback loop.

6. 06Physical AI

   Applies learned perception, prediction, and control in embodied systems.

7. 07Digital twin

   Maintains an operational digital representation linked to a physical system.

• LLM informs VLM.
• VLM informs World Model.
• World Model predicts for Planner.
• Planner plans with Agent.
• Agent acts through Physical AI.
• Digital Twin feeds back to World Model.