CourseTransformers: From Attention to Trained, Served and Interpreted Models · Module 10: Efficiency and Long Context · part 70 of 111

Part 70 · Module 10: Efficiency and Long Context

Module 10: Efficiency and Long Context

1 min read·13 Oct 2026

What you need first

Module 8 and its lab (tinyformer-lab at v0.8.0, or v0.9.0 after Module 9's lab; the lab can rebuild a stand-in)
Module 2, Lesson 5 (attention cost and the KV cache by hand) and Module 4 (rotary positions and context extension)
Module 6: the base recipe and the seed noise floor
No GPU, no API key; Lesson 3 downloads published config files from the Hugging Face Hub

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