NLP & Language · intermediate · concept 83 of 176

🗣️Context Window & Long Context

The maximum number of tokens an LLM can process at once. A window of roughly one million tokens is now standard at the frontier, with a few models advertising ten million. That is enough for entire codebases or books, but effective recall degrades well before the advertised limit, so the usable window is usually much smaller than the number on the spec sheet. Long context has a known failure mode: models attend best to the start and end of the window, the lost-in-the-middle effect.