Structured Output

Techniques

Simple Definition

Constraining an AI model to generate responses in a specific format (JSON, XML, etc.) for reliable programmatic processing.

Full Explanation

By default, LLMs generate free-form text. Structured outputs force the model to produce valid JSON (or other formats) matching a specified schema. This is essential for production AI applications — you can reliably extract data, fill databases, and process AI outputs programmatically without fragile string parsing. OpenAI, Anthropic, and Google all offer structured output modes.

Related Terms

Tool Use / Function Calling

The ability for AI models to call external functions, APIs, or tools to retrieve information or take actions beyond their training data.

Last verified: 2026-03-30← Back to Glossary