Summary
Unit 8 of 8 · Module 4: Develop generative AI apps that use tools
Read this lesson
Stitched together
And that's Module 4. With tools, a model stops only talking about the world and starts doing useful, grounded things in it. You added tools to your Responses API requests, and each one gave the model a new skill. Code interpreter let it write and run Python, for sums, data crunching, and trying again until a problem is solved. Web search brought in fresh information from outside, with answers that point to where they came from. File search let it answer from your own indexed documents. And the function tool let your app run its own business logic, then hand the result back to the model. Every tool follows the same pattern: describe it in your request, let the model decide when it's needed, get the tool's result back to the model, which Azure does for you with the built-in tools and your app does for functions, and check the answer is correct and safe.
The next patch
From here, you can combine these tools into more capable assistants, and grow them into full agentic solutions, where instructions, tools, and orchestration are saved and managed for you. To learn more, OpenAI's developer guides on tools and function calling are worth a read. Next up, you'll learn how to get better performance out of generative AI models in Microsoft Foundry.
Facts from Microsoft's documentation: Use the Azure OpenAI Responses API; Web search with the Responses API; File search tool for Microsoft Foundry agents; How to use function calling with Microsoft Foundry Models (CC BY 4.0), rewritten in Loop's own words and examples. Narration, animation, and quiz by Loop. Microsoft, Azure, and Microsoft Foundry are trademarks of Microsoft. Loop is independent and not affiliated with or endorsed by Microsoft.