Fructose
by Community
Fructose is a python package to create a dependable, strongly-typed interface around an LLM call. ![GitHub Repo stars](https://img.shields.io/github/stars/bananaml/fructose?style=s
OSS
Fructose
Added 1 June 2026
Overview
Fructose is a Python package that uses decorators and type hints to create a strongly-typed interface for LLM calls. It wraps any function with a typed signature, converting it into a structured LLM invocation that returns predictable outputs.
Best for
Best for
Python developers who want reliable, typed interfaces for LLM calls in their applications
Use cases
- Defining typed LLM functions for data extraction and classification
- Integrating LLM calls into existing Python codebases with type safety
- Building reliable LLM pipelines that produce structured outputs
Notes
Fructose is a Python package that uses decorators and type hints to create a strongly-typed interface for LLM calls. It wraps any function with a typed signature, converting it into a structured LLM invocation that returns predictable outputs.
750 stars on GitHub. Last updated 2024-04-17. Licensed Apache-2.0.
Use cases
- Defining typed LLM functions for data extraction and classification
- Integrating LLM calls into existing Python codebases with type safety
- Building reliable LLM pipelines that produce structured outputs
Pros
- Strong typing ensures predictable and verifiable outputs
- Simple decorator-based API minimizes boilerplate
- Works seamlessly with Python type checkers for early error detection
Cons
- Limited to Python; no direct support for other languages
- May restrict flexibility for complex or dynamic prompting patterns
- Depends on the LLM’s ability to correctly follow structured output instructions
Indexed from awesome-langchain and enriched against its public facts.
Pros
- Strong typing ensures predictable and verifiable outputs
- Simple decorator-based API minimizes boilerplate
- Works seamlessly with Python type checkers for early error detection
Cons
- Limited to Python; no direct support for other languages
- May restrict flexibility for complex or dynamic prompting patterns
- Depends on the LLM's ability to correctly follow structured output instructions
Pairs with
Other entries in the index that connect to this one. Click through to see the chain.
Get the free Developer’s Field Guide
A 27-page field guide to the AI coding workflow with Claude. Claude Code, MCP servers, the prompt patterns that work, and what to delegate. Free.
Enter your work email. We send it straight over, plus a few short notes worth knowing. Unsubscribe any time.