Context Window

Definition

The maximum amount of text a language model can process at once, measured in tokens. GPT-4 Turbo has a 128K context window; Claude can handle 200K+. Larger context windows let models work with longer documents but use more memory and compute. A key differentiator between models.

How It Works

The context window is the maximum amount of text a language model can process at once — both the input you give it and the output it generates. Think of it as the model's working memory. Anything outside the context window simply doesn't exist for the model during that conversation.

Context windows have grown dramatically. GPT-3 had about 4,000 tokens (~3,000 words). GPT-4 Turbo expanded to 128K tokens. Claude can handle 200K tokens — roughly a 500-page book. Google's Gemini 1.5 Pro pushed to 1 million tokens. This expansion matters because longer context means you can feed the model entire documents, codebases, or conversation histories.

But bigger doesn't always mean better in practice. Models can struggle with information buried in the middle of very long contexts — a phenomenon researchers call "lost in the middle." They tend to pay more attention to the beginning and end. There's also the cost factor: processing longer contexts requires more compute, which means higher API costs. Smart applications use techniques like RAG to pull in only the most relevant information rather than dumping everything into the context window.