About Guardrails#
Use NeMo Guardrails to apply safety checks and content moderation to large language model (LLM) applications. The service evaluates user inputs and model outputs against configurable guardrail policies, supports dedicated task models (such as content safety or topic control), and routes inference requests through the Inference Gateway service.
Guardrails configurations define which checks run, which models perform the checks, and how blocked content is handled. Your application runs inference using the Guardrails OpenAI-compatible inference endpoints, and NeMo Guardrails orchestrates the checks before returning the final response.
Tutorials#
See Guardrails Architecture for an overview of the Guardrails architecture and request flow.
These tutorials walk you through common guardrail tasks using NeMo Guardrails.
Tutorials for common guardrail tasks, using the Helm installation to prepare for production setup.
API Usage Guides#
These guides cover common guardrail operations using the NeMo Guardrails APIs, either through the NeMo Platform Python SDK or directly via the REST API.
Manage guardrail configurations.
Check content with input and output guardrails.
Run inference and apply safety checks.