Guardrails AI
Add input/output validation and safety rails to LLM calls
Verdict
Cleanest Python API for defining validators on LLM inputs and outputs. RAIL spec and Pydantic-based guards. Retry on failure is well-designed. Server mode needed for prod latency.
Other Guardrails & Safety
- Anthropic Constitutional AIStable
Principle-based self-improvement for harmlessness and helpfulness
- Azure AI Content SafetyProduction
Microsoft's content moderation API for text and images
- Lakera GuardStable
Real-time prompt injection and jailbreak protection API
- Llama Guard 3Experimental
Meta's fine-tuned safety classifier for prompt and response screening
- Microsoft PresidioProduction
Data protection and anonymization for PII in LLM pipelines
- NeMo GuardrailsExperimental
NVIDIA toolkit for programmable guardrails via Colang language