What is Token Context Window?
The maximum number of tokens a language model can process in a single request, encompassing both input and output.
The token context window defines the upper boundary of information a large language model can consider at once. Every piece of text—system prompts, user messages, retrieved documents, tool call results, and the model's own generated output—must fit within this window. When the total token count exceeds the limit, older content is typically truncated or must be summarized. Understanding context window constraints is critical for designing effective agents, as it influences decisions about retrieval strategies, conversation management, skill decomposition, and when to delegate work to sub-agents.