Skip to main content
Ctrl+K
NVIDIA NeMo Guardrails - Home NVIDIA NeMo Guardrails - Home

NVIDIA NeMo Guardrails

NVIDIA NeMo Guardrails - Home NVIDIA NeMo Guardrails - Home

NVIDIA NeMo Guardrails

Table of Contents

NVIDIA NeMo Guardrails

  • About NeMo Guardrails
  • Installation Guide

Getting Started

  • Hello World
  • Core Colang Concepts
  • Demo Use Case
  • Input Rails
  • Output Rails
  • Topical Rails
  • Retrieval-Augmented Generation

Colang 2.0

  • Overview
  • What’s Changed
  • Getting Started
    • Hello World
    • Dialog Rails
    • Multimodal Rails
    • Input Rails
    • Interaction Loop
    • LLM Flows
    • Recommended Next Steps
  • Language Reference
    • Introduction
    • Event Generation & Matching
    • Working with Actions
    • Defining Flows
    • Working with Variables & Expressions
    • Flow control
    • Colang Standard Library (CSL)
      • Fundamental Core Flows (core.co)
      • Timing Flows (timing.co)
      • LLM Flows (llm.co)
      • Interactive Avatar Modality Flows (avatars.co)
      • Guardrail Flows (guardrails.co)
      • User Attention Flows (attention.co)
    • Make use of Large Language Models (LLM)
    • More on Flows
    • Python Actions
    • Development and Debugging

User Guides

  • Configuration Guide
  • Guardrails Library
  • Guardrails Process
  • Colang Guide
  • LLM Support
  • Python API
  • CLI
  • Server Guide
  • LangChain
    • LangChain Integration
    • RunnableRails
    • Chain-With-Guardrails
      • Chain with Guardrails
    • Runnable-As-Action
      • Runnable as Action
  • Detailed Logging
    • Output Variables
  • Jailbreak Detection Heuristics
    • Using Jailbreak Detection Heuristics
  • NVIDIA AI Endpoints
    • Using LLMs hosted on NVIDIA API Catalog
  • Vertex AI
    • Using LLMs hosted on Vertex AI
  • Multi Config API
    • Multi-config API
  • Migrating from Colang 1 to Colang 2

Security

  • Security Guidelines
  • Red Teaming

Evaluation

  • Guardrails Evaluation
  • LLM Vulnerability Scanning

Advanced User Guides

  • Generation Options
  • Prompt Customization
  • Embedding Search Providers
  • NeMo Guardrails with Docker
  • Streaming
  • AlignScore Deployment
  • Extract User-provided Values
  • Bot Message Instructions
  • Event-based API
  • Self-hosting Llama Guard using vLLM
  • Nested AsyncIO Loop
  • Vertex AI Setup
  • Llama 3.1 NemoGuard 8B ContentSafety Deployment
  • Llama 3.1 NemoGuard 8B Topic Control Deployment
  • Jailbreak Detection Deployment
  • Blueprint with NemoGuard NIMs

Reference

  • Architecture
    • Architecture Guide
  • Configuration File Reference
  • Glossary
  • Frequently Asked Questions (FAQ)
  • Getting Started
  • Recommended Next Steps

Recommended Next Steps#

The Colang 2.0 getting started guide introduced you to a basic Hello World example, and then goes into Dialog Rails, Input Rails, Multimodal Rails, and other Colang 2.0 concepts like Interaction Loop and LLM Flows.

This only scratches the surface of what can be achieved with Colang 2.0.

If you are an experienced developer and want to learn about the syntax and the various features in details, we recommend going through the Language Reference documentation.

Version 0.10.0 of NeMo Guardrails will add more examples including RAG and agents. Also it will bring support for the Guardrails Library, which will enable you to use any of the existing guardrails similar to Colang 1.0.

previous

LLM Flows

next

Language Reference

NVIDIA NVIDIA
Privacy Policy | Manage My Privacy | Do Not Sell or Share My Data | Terms of Service | Accessibility | Corporate Policies | Product Security | Contact

Copyright © 2023-2025, NVIDIA Corporation.