epochs

Learn · Module 12 · Inference and serving · Lesson 5 of 6

What a request costs

AI costs are easy to estimate on paper and easy to be surprised by in production. Do the arithmetic before you launch.

3 minute read

In short
  • Cost per request = input tokens × input price + output tokens × output price.
  • Model choice is the biggest lever; small models can be over ten times cheaper.
  • Retries, agents, long histories and retrieved text multiply costs quietly.

This lesson is part of Pro

The first two lessons of every module are free. Pro opens every lesson and every project.

See Pro plansFrom ₹119 a month. No auto-renewal.