Introduction

Belso (Better LLMs Structured Outputs) is a Python library designed to simplify working with structured outputs from Large Language Models (LLMs). It provides a unified interface for validating, translating, and processing structured data across multiple LLM providers.

Key Features

• Multi-Provider Support: Seamlessly work with outputs from:

  • OpenAI

  • Anthropic

  • Google AI

  • Mistral AI

  • Hugging Face

  • Ollama

  • LangChain

• Format Translation: Convert between different schema formats

• Nested Schemas: Easily define and validate deeply nested structures

• Export Formats: JSON, YAML, XML support out of the box

• Simple API: Developer-friendly and extensible