Release history
Guardrails releases
NeMo Guardrails is an open-source toolkit for easily adding programmable guardrails to LLM-based conversational systems.
All releases
11 shown
Config change
v0.22.0
Breaking risk
Auth
Usage reporting, LangChain optional, IORails enhancements
v0.21.0
Breaking risk
Breaking changes
- Removed stream_usage from streaming metadata capture
Notable features
- IORails Input/Output rail engine with parallel execution
- check_async method for standalone rail validation
- OpenAI-compatible server with v1/models endpoint
v0.20.0
Breaking risk
Breaking changes
- Removed streaming field from configuration
Notable features
- Nemotron reasoning-capable content safety models
- GLiNER for open-source PII detection
- Multilingual refusal messages in content safety rails
v0.19.0
Breaking risk
Notable features
- LangChain 1.x compatibility
- Content blocks API support for reasoning traces and tool calls
v0.18.0
Breaking risk
Breaking changes
- Reasoning traces extracted to separate field instead of prepended to message
- stream_async now raises error if output rail streaming disabled
- Replaced reasoning trace extraction with LangChain additional_kwargs
Notable features
- BotThinking events for processing reasoning traces
- Embedding providers: Azure OpenAI, Cohere, Google
- In-memory LFU caches for content-safety, topic-control, jailbreak models
v0.17.0
New feature
Notable features
- LangGraph integration for multi-agent workflows
- Tool calling with passthrough and validation
- RunnableRails rewrite with full Runnable protocol support
v0.16.0
New feature
Notable features
- OpenTelemetry semantic conventions for tracing
- GuardrailsAI and Pangea AI Guard integrations
- KV cache reuse for LLM-based NemoGuard models
v0.15.0
Breaking risk
Breaking changes
- Tracing updated to use OpenTelemetry API
Notable features
- Parallel rails execution in Colang 1.0
- Parallel streaming output rails execution
- Fixed streaming word concatenation and token usage tracking
v0.14.1
Bug fix
Notable features
- Direct API key configuration for jailbreak detection
v0.14.0
Breaking risk
Breaking changes
- Reasoning trace preservation and handling restructured
- Deprecated return_context argument removed
- Renamed remove_thinking_traces to remove_reasoning_traces
Security fixes
- YARA-based code injection detection guardrail
Notable features
- Reasoning model support for Nemotron and DeepSeek-r1
- YARA-based code injection detection guardrail
- Clavata community integration for content moderation