Lab
A demo is the beginning.
Explore what changes when an AI application meets the real world.
◇ Educational simulation. Not a running Kong deployment or a benchmark; numbers are illustrative.
Introduce challenges
Turn on one or several. Each adds a problem and the control that addresses it.
Request path
Select any component for a short explanation.
- AI provider
About this simulation
This is an educational simulation that runs entirely in your browser. It isn't a running Kong deployment, it doesn't call any AI provider, and its numbers aren't benchmarks.
Token counts, budgets and provider limits are illustrative values chosen to make each decision visible. Real systems measure tokens with the provider's tokenizer, enforce budgets across many gateway nodes, and need far more careful detection of sensitive data than the simple pattern used here.
Read more
API Gateways for AI Workloads→How API gateways are evolving to handle the unique demands of LLM traffic — streaming, token limits, and semantic caching.