Choosing an LLM and AI stack

Stage 1 · Lesson 3 of 6

The cheapest model that is good enough

Start from the task, not the model

The goal is not the smartest model. It is the cheapest and fastest model that clears your quality bar for this task. A model that is 2% better but ten times the price is usually the wrong choice for sorting support tickets, and the right choice for reviewing a legal contract.

A sensible order is: prove the feature works with a strong model, write down what "good enough" means as , then try smaller models against the same cases and while quality holds. Many products end up using two or three models: a small one for most requests and a strong one for the hard ones. Keep this idea in mind through every later stage.

A question to check yourself, an AI practice session and 2 cheat sheet entries come with the full roadmap. Unlock

Resources for this stage

Start with· follow one of these

Practise

  • Try the same prompt on several Gemini tiers in the browser and feel the trade-off yourself

Keep handy

Also worth your time

Mark this stage as not done?