Llama 3.1 NemoGuard 8B ContentSafety NIM

Type: Microservice Tags: NVIDIA, NIM, NemoGuard, content safety, moderation, guardrails, Llama 3.1, AI safety Related: NVIDIA-NemoGuard-NIMs, NeMo-Guardrails, NeMo-Platform, NVIDIA-NIM, NIM-for-Large-Language-Models, Llama-3.1-Nemotron-Safety-Guard-8B-NIM, Llama-3.1-NemoGuard-8B-TopicControl-NIM, NVIDIA-NemoGuard-JailbreakDetect-NIM, Nemotron, NVIDIA-AI-Enterprise Sources: https://docs.nvidia.com/nim/llama-3-1-nemoguard-8b-contentsafety/latest/index.html, https://docs.nvidia.com/nim/llama-3-1-nemoguard-8b-contentsafety/latest/integrate-guardrails.html, https://docs.nvidia.com/nemo/guardrails/latest/configure-rails/guardrail-catalog/content-safety.html Last Updated: 2026-04-29

Summary

Llama 3.1 NemoGuard 8B ContentSafety NIM is a GPU-accelerated content-moderation NIM for trustworthy LLM applications. Current NVIDIA docs describe it as detecting harmful content in user messages or bot responses.

Detail

Purpose

ContentSafety NIM gives applications a deployable safety classifier for input and output moderation. It can be called directly through NIM endpoints or integrated into NeMo-Guardrails content-safety rail flows.

Current scope

  • Detects harmful content in user messages and bot responses.
  • Trained by NVIDIA from a Llama 3.1 8B Instruct base model.
  • Uses the Nemotron Content Safety Dataset V2 lineage, formerly Aegis AI Content Safety Dataset 2.0.
  • Exposes standard NIM LLM endpoints such as completions and chat completions.
  • NeMo Guardrails integrates through LangChain ChatNVIDIA and a content_safety model configuration.

NVIDIA context

ContentSafety NIM is the earlier NemoGuard content-safety NIM lineage. The newer Llama-3.1-Nemotron-Safety-Guard-8B-NIM covers multilingual Safety Guard V3; both remain useful query targets for NVIDIA content moderation.

Connections

Source Excerpts

  • NVIDIA docs describe ContentSafety NIM as a GPU-accelerated LLM model that performs content moderation.
  • The docs state that the model detects harmful content in user messages or bot responses.

Resources