LlamaGuard

ObservabilityGuardrails

Meta LLM-based input/output safeguard model. Classifies safety risks in both user prompts and agent responses in real time.

Visit website View on GitHub
Built with
PythonPyTorchHuggingFace
Share this part