Model Caching

Storing the results of common AI requests so you can serve them faster next time — like saving frequently ordered dishes at a restaurant so they can be served instantly. It cuts down latency and saves compute costs. But if the underlying data changes, you might serve stale answers.