Agents are the core components that interact with users and perform tasks. Agent Versions are specific implementations of agents with defined behaviors, prompts, tools, and configurations. Together, they allow you to track changes in behavior over time while keeping a stable logical identity for your assistant.

An agent represents a logical assistant or capability in your system. It defines the high-level purpose, such as "Support Bot" or "Document Analyzer." Agents are long-lived and typically remain stable, while their versions evolve over time.

An agent version represents a specific configuration of an agent at a point in time. This includes the model, system prompt, tool configuration, and any other behavioral settings. Creating new versions allows you to iterate safely and compare performance across changes.