Glossary · Phase 11 · Professional practice
Model routing
Sending each request to a suitable model, such as easy ones to a small cheap model and hard ones to a large one.
Try it now
Requests × tokens × price, using the Opus and Haiku prices from Phase 1. Send easy requests to the small model and watch the bill drop.
Learn it properly
Designing AI systems · Latency, cost, routing, fallbacks and guardrails for features built on models. About 40 minutes.
Other terms from the same lesson
- circuit breaker: A guard that stops calling a failing service for a while after repeated errors, then lets one trial call through.
Never trained an AI?
Teach a small one to write, right in your browser. It takes about 20 seconds and needs no account.