The maximum amount of text (measured in tokens) that an AI model can process in a single conversation or request. Like the size of a desk - determines how many documents you can spread out at once.
GPT-4 Turbo has a 128K token context window, allowing it to analyze an entire book or codebase in one conversation.