NVIDIA NeMo Guardrails

NVIDIA NeMo Guardrails lets developers add programmable safety and security controls to LLM applications and tool-using agents.

Verified 11 Nov 2026

Facts

Source per value

Self-host or private cloud The vendor documents running the product on your own servers, in your own cloud or VPC, or air-gapped. Derived from the deployment text on the same source.: Yes 1
Runtime guardrails, as documented What the vendor documents for this capability.: Input, retrieval, dialog, execution, and output rails support policy enforcement. 1
Agent security, as documented What the vendor documents for this capability.: Validates tool inputs and outputs around tool invocation. 1
Integrations and platforms Named clouds, models, development tools, protocols or security tools the product supports.: Integrates with LangGraph for multi-agent safety. 1
Deployment Documented hosting and installation options.: Self-managed Python library or microservice deployed on Kubernetes with Helm. 1
Licensing Open-source status or paid packaging, where documented.: NeMo Guardrails library is an open-source Python package. 1

A10 Networks TrojAI: TrojAI by A10 Networks provides AI runtime defense, agent-led red teaming, and MCP governance for enterprise agents, applications, and models.
Arthur: Arthur offers enterprise agent security and governance with discovery, behavioral analytics, runtime guardrails, and audit evidence.
Aurascape: Aurascape secures enterprise AI usage and AI agents with discovery, data protection, adversarial testing, and runtime controls.
Check Point AI Security: Check Point sells AI security controls for employee usage, applications, and autonomous agents, including runtime protection and adversarial assessment.

Tell us what you need once. We send it to matching vendors and they reply to you directly.