guidance
Make your LLM outputs fit the format you need.
What it does
Guides developers in writing constrained LLM generation code that produces structured outputs and multi-step workflows.
When to use it
Use when LLM output must follow regex patterns, grammars, or structured formats, or when building multi-step generation workflows.
How to use it
It changes how the agent writes LLM generation code by using Guidance constraints, fixed selections, and reusable generation functions.
What you provide
- An existing project
Uses
What you need · 5
The generation examples use the guidance Python library.
The examples and generation workflows are written in Python; no runtime version is specified.
Enables the Anthropic API backend as an alternative to local models; the key can be read from the environment or supplied explicitly through api_key.
Enables the OpenAI API backend as an alternative to local models; the key can be read from the environment or supplied explicitly through api_key.
Access · 0
No additional actions listed.