The maximum amount of text (measured in tokens) an AI model can consider at once when generating a response.
A model's context window limits how much conversation history, document content, or instructions it can 'see' at one time — a key constraint when designing RAG and agent systems.
