GitHub 저장소 URL
https://github.com/567-labs/instructor
분류 (모듈)
platform-mlops
카테고리
agent-orchestration
설명 (선택)
LLM에서 구조화된 출력(structured outputs)을 안정적으로 얻기 위한 라이브러리 Instructor. Pydantic 모델로 응답 스키마를 정의하고 검증·재시도를 자동화하며 다양한 LLM 제공자를 지원합니다.
체크리스트
추가 컨텍스트 (검토용)
- Stars: 13,109
- Forks: 1,071
- License: MIT
- Language: Python
- Last Pushed: 2026-06-05
- koreanSupport: 미확인 (영어 문서 중심)
- 제안 태그: structured-outputs, pydantic, llm, validation, python-library
- 분류 근거: LLM 구조화 출력을 위한 앱 개발 라이브러리이므로 platform-mlops / agent-orchestration에 배치.
- 검토 참고: 저장소가 jxnl/instructor → 567-labs/instructor로 이동(리디렉트). 양쪽 경로 모두 미등록 확인. LLM 앱 프레임워크/SDK 군집(Mirascope 등)과 카테고리 검토 권장.
GitHub 저장소 URL
https://github.com/567-labs/instructor
분류 (모듈)
platform-mlops
카테고리
agent-orchestration
설명 (선택)
LLM에서 구조화된 출력(structured outputs)을 안정적으로 얻기 위한 라이브러리 Instructor. Pydantic 모델로 응답 스키마를 정의하고 검증·재시도를 자동화하며 다양한 LLM 제공자를 지원합니다.
체크리스트
추가 컨텍스트 (검토용)