Modern AI platforms give us lots of models and that can make things difficult. Different costs, latencies, capabilities, and quality trade-offs mean that hard-coding a single model into your application is rarely the best choice.

That's where the Model Router in Microsoft Foundry comes in. In this session, you’ll see how routing decisions can be based on factors like task complexity, performance requirements, and cost constraints—without pushing that logic into your application code.

If you’re building AI-powered applications and don’t want to bet everything on a single model, this session will show you a smarter way forward.