Small Models, Big .NET Energy
In this session from dotnet, demonstrations of Foundry Local show what small AI models can do on a developer’s own hardware. The discussion takes a C# perspective on choosing when local execution is a useful fit.
Foundry Local and small models
Session BRK037 focuses on practical local AI demonstrations rather than model size alone. It explores what is possible without a cloud connection and where local AI shines.
For .NET developers, the central consideration is balancing three factors:
- Capability: What the model can accomplish for the task.
- Speed: How quickly it runs on local hardware.
- Cost: How cost influences the choice of model and execution approach.
The session uses these trade-offs to frame where small models and Foundry Local fit into C# development.