Chapter 02

Cost, quotas, and rate limits

The real economics of running LLM-based systems: understanding the true cost per call, setting budgets and caps that protect you, and using caching to cut spend without hurting quality.