NVIDIA NeMo Guardrails Toolkit Developer Guide#

The NeMo Guardrails toolkit is an open-source Python package for adding programmable guardrails to LLM-based applications. It intercepts inputs and outputs, applies configurable safety checks, and blocks or modifies content based on defined policies.

About the NeMo Guardrails Toolkit#

Learn about the toolkit and its capabilities in the following sections.

Overview

Learn about the NeMo Guardrails toolkit and its capabilities.

Overview of the NeMo Guardrails Toolkit
Use Cases

Browse the different use cases of the NeMo Guardrails toolkit.

Use Cases
How It Works

Learn how the NeMo Guardrails toolkit works.

How It Works
Supported LLMs and Providers

Browse the LLMs and their providers supported by the toolkit.

Supported LLMs

Get Started#

Follow these steps to start using the NeMo Guardrails toolkit.

Install

Install the toolkit with pip and set up your environment.

Installation Guide
Tutorials

Follow hands-on tutorials to build your first guardrails configuration.

Tutorials

Next Steps#

Once you’ve completed the get-started tutorials, explore the following areas to deepen your understanding.

Configure Rails

Learn to write config.yml, Colang flows, and custom actions.

Configuration Overview
Run Rails

Use the Python SDK and understand core classes like RailsConfig and LLMRails.

Run Rails
Evaluate

Evaluate the performance of the rails.

Guardrails Evaluation
Observability

Monitor and troubleshoot your guardrails applications.

Logging and Debugging
Deploy

Deploy your guardrails using the toolkit’s local server, Docker, or as a production microservice.

Deployment Options
Integrate

Connect with LangChain, LangGraph, and other frameworks.

LangChain Integration