context window
also: token limit, context length
The amount of text (measured in tokens) that a language model or AI system can process and reference at one time. It defines the maximum length of conversation history or input the model can consider.
A context window is a fixed limit on how much information an AI language model can "see" or process in a single request. Think of it like the model's working memory—everything outside this window is forgotten or unavailable.
For example, if a model has a 4,000 token context window and you send it a 5,000 token document plus a question, the model cannot reference the entire document; it can only work with the most recent 4,000 tokens (roughly 3,000 words). This is why long conversations sometimes lose earlier details.
In practical Linux and DevOps use, larger context windows in AI tools are valuable for analyzing logs, reviewing code, or asking questions about large configuration files—the AI can keep more relevant information in mind while helping you troubleshoot or write scripts.