CourseLarge Language Models: The Complete Course · Module 2: Tokens, Context, and Cost · part 16 of 103

Part 16 · Module 2: Tokens, Context, and Cost

Lesson 2: What fits in the window, and what the model uses

21 min read·13 Oct 2026

Module 2 · Lesson 2 · About 60 minutes · Needs: Module 1 (TinyLM, SmolLM2) and what a token is (Lesson 1)

The situation

It is Wednesday 17 June 2026, week 3. Ana Duarte's chat about the duplicate charge on invoice INV-2026-004512 has grown: Slack, an export, and now "Any news on my refund?". Every reply resends the whole conversation. Tomasz asks how much of the model's window one request uses, and what happens when it does not fit. Maya asks: if the invoice id is in there, will the model use it? (Quillpath and its data are made up for this course.)

Every part of Ana's request (blue, history in amber because it grows) shares one window with the reply the model has not written yet (grey).

The rest of this course is yours to keep

This course is bought on its own, once, and stays readable afterwards, including the parts added to it later.

Buy this course · ₹399 See what is inside