⚙️ Under the hood
Context window
The amount of text (in tokens) a model can consider at once: system instructions, chat history, files and your question.
💭 Imagine …
The model's desk. What's on it can be used. When it's full, old things fall off or must be summarised.
🔎 Example
Modern models often handle hundreds of thousands of tokens, entire books. Still, details in long texts are sometimes missed.
🎯 Why it matters
The model has no memory between chats unless the tool deliberately stores it. What matters belongs in the context.
🎓 Prefer step by step?
The AI Literacy course: 6 short modules with quizzes, a solid foundation in about 60 minutes.