CourseTransformers: From Attention to Trained, Served and Interpreted Models · Module 11: Inference and Serving · part 77 of 111

Part 77 · Module 11: Inference and Serving

Module 11: Inference and Serving

2 min read·13 Oct 2026

What you need first

Module 10 and its lab (tinyformer-lab at v0.10.0; the lab can rebuild a stand-in, with Module 9's files)
Module 5 (the model and its causal mask) and Module 6 (the base recipe and the seed noise floor)
No GPU, no API key; Lesson 5 downloads GPT-2 small, and engine commands (vLLM, SGLang, llama.cpp) are shown, not run

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