instructor - Structured LLM outputs with Pydantic validation
Extracts and validates structured data from LLM responses, retries failed validations, and streams partial results across supported providers.
Tags
Updated: 2026-10-02Capabilities
Typical Inputs
Typical Outputs
What this skill does
- Extract structured data
- Validate Pydantic outputs
- Retry failed extractions
- Parse typed JSON
- Stream partial results
- Support multiple providers
Inputs
- Input text
- Pydantic response models
- Provider configuration
- API keys
- LLM model names
Outputs
- Validated structured objects
- Partial result updates
- Streamed iterable items
- LLM response messages
Requirements
- Python packages: instructor, pydantic, openai, anthropic
- Supported operating system
- LLM provider access
- Provider API credentials