Core concepts
Context Window
The maximum amount of text (tokens) an AI model can read and remember at once during a single conversation.
01 ——
In plain English
The context window is the "working memory" of an AI model — it's the total amount of text the model can process in one go, including the conversation history, instructions, and any documents you've shared.
It's measured in tokens (roughly ¾ of a word each). A model with a 200,000-token context window can hold roughly 150,000 words at once — about two full novels.
Why it matters:
- Short context = the model forgets earlier parts of a long conversation
- Long context = the model can reason over entire codebases or research papers at once
Models like Claude and GPT-4 have pushed context windows from 4k tokens in 2022 to 1M+ tokens in 2025.
02 ——
Related terms
LLM
Large Language Model — the type of AI behind tools like ChatGPT and Claude, trained to understand and generate text.
Tokens
The basic units of text that AI models read and write — roughly ¾ of a word each. Models are priced and limited by token count.
RAG
Retrieval-Augmented Generation — a technique that gives an AI model access to external documents before it answers, so it can cite real, up-to-date sources.
Prompt Caching
A feature that stores parts of a prompt the model has already processed, making repeat or follow-up requests much faster and cheaper.
Back to glossaryLast reviewed May 2026