This repository provides sample applications demonstrating OpenAI's Structured Outputs feature, designed to help developers reliably integrate AI models into deterministic workflows by enforcing JSON schema adherence for responses and tool calls. The target audience includes developers building applications with OpenAI models, particularly those using NextJS, who need predictable and structured AI-generated data.

How It Works

The core approach leverages OpenAI's API to generate outputs that conform to a predefined JSON schema. This ensures that model responses, including tool calls, are consistently formatted, simplifying data parsing and integration into downstream processes. The sample apps showcase this by using Structured Outputs for tasks like extracting information into a structured format, dynamically generating UI components, and managing multi-turn conversations with tool integration.

Quick Start & Requirements

• Install: Clone the repository (git clone https://github.com/openai/structured-outputs-samples.git). Navigate to individual app folders and follow their specific READMEs for local execution.
• Prerequisites: OpenAI API key.
• Resources: Refer to Structured Outputs documentation, Function Calling documentation, and the Introduction to Structured Outputs cookbook.

Highlighted Details

• Demonstrates Structured Outputs for resume extraction.
• Showcases generative UI component creation.
• Includes a conversational assistant with multi-turn and tool-calling capabilities.
• Focuses on practical application development using NextJS.

Maintenance & Community

Contributions via issues or PRs are welcome, though review is not guaranteed.

Licensing & Compatibility

Licensed under the MIT License, permitting commercial use and integration with closed-source projects.

Limitations & Caveats

The repository contains sample applications and may not represent a production-ready framework. Specific app functionalities and their robustness depend on the individual app's implementation.