Skip to navigation

Install the NeMo Guardrails Library

View as MarkdownOpen in Claude

Follow these steps to install the NeMo Guardrails library.

Prerequisites

Verify your system meets the following requirements before installation.

RequirementDetails
Operating SystemWindows, Linux, MacOS
Python3.11, 3.12, or 3.13
Hardware1 CPU with 4GB RAM. The NeMo Guardrails library runs on CPU. External models may require GPUs, which may be deployed separately to the library

Starting with version 0.25.0, the NVIDIA NeMo Guardrails library requires Python 3.11 through 3.13. Python 3.10 reached end of life on October 1, 2026. Upgrade your Python environment before upgrading the library.

Quick Start

Use the following steps to install the NeMo Guardrails library in a virtual environment.

  1. Create and activate a virtual environment:

    python3 -m venv .venv
    source .venv/bin/activate
  2. Install the NeMo Guardrails library. Set NVIDIA_API_KEY to your personal API key generated on build.nvidia.com.

    pip install nemoguardrails
  3. Set up an environment variable for your NVIDIA API key.

    export NVIDIA_API_KEY="your-nvidia-api-key"

    This is required to access NVIDIA-hosted models on build.nvidia.com. The tutorials and example configurations (examples/configs) in this library include configurations that use NVIDIA-hosted models.

Verify the Installation

Run these commands in the virtual environment where you installed the library:

python -c "import nemoguardrails; print(nemoguardrails.__version__)"
nemoguardrails --help

The first command prints the installed package version. The second lists the CLI commands. These checks do not call a model or require an API key. Follow a tutorial to verify inference with your selected models.

Get the Example Files

Installing with pip installs the Python package. To follow instructions that use repository paths such as examples/configs/, clone the repository and run those commands from its root:

git clone https://github.com/NVIDIA-NeMo/Guardrails.git
cd Guardrails

For a released package, check out its matching release tag before using the examples. For example, use git checkout v<installed-version>, replacing <installed-version> with the version printed above. The develop branch contains development examples that may require a newer package.

Alternative Installation Methods

Install the NeMo Guardrails library from source using pip or uv. Choose this method if you want to contribute to the library or use the latest development version.

git clone https://github.com/NVIDIA-NeMo/Guardrails.git nemoguardrails
cd nemoguardrails
python -m venv .venv
source .venv/bin/activate
pip install -e .

Extra Dependencies

You can install the NeMo Guardrails library with optional extra packages to add useful functionalities. The table below shows a comprehensive list.

ExtraDescription
serverGuardrails API server dependencies (aiofiles for async file handling, openai for API schemas). FastAPI is a core dependency. Required to run nemoguardrails server and the Guardrails hooks server, nemoguardrails hooks-server.
sddSensitive data detection using Presidio
evalPolicy-based evaluation tools for testing guardrails
tracingOpenTelemetry tracing support
gcpGoogle Cloud Platform language services
jailbreakYARA-based jailbreak detection heuristics
multilingualLanguage detection for multilingual content
chat-uiChainlit chat UI dependencies
celCommon Expression Language (CEL) rail support using common-expression-language (Python 3.11 and later)
nest-asyncioOptional compatibility support for synchronous calls inside a running event loop. Automatic patching will be deprecated in 0.26. See Nested AsyncIO Loop.
allAll optional packages

Starting with 0.25, the default installation does not require nest-asyncio. Use async APIs in notebooks and async applications. Existing examples that import nest_asyncio require pip install 'nemoguardrails[nest-asyncio]'. The chat-ui and all extras still install nest-asyncio through Chainlit. If you need an installation without that package, avoid those extras and verify your application’s other dependencies.

Some features such as AlignScore have additional requirements. See the feature documentation for details.

Docker

You can run the NeMo Guardrails library in a Docker container. For containerized deployment, see NeMo Guardrails with Docker.