NeMo Guardrails

No entries match these filters. Clear filters

1. v0.24.0 Aug 26, 2026 · issue 058

   NeMo Guardrails v0.24.0 adds RailOutcome, rail manifests, F5 integration, health endpoints, and expands IORails to 59 built-in rails.

   └──▷ GET THIS VERSION

   copy

   $ git clone --branch v0.24.0 https://github.com/NVIDIA-NeMo/Guardrails.git
   # already have the repo? check out this version:
   $ git checkout v0.24.0

   └──▷ TRY IT
   Check server health before routing traffic to a NeMo Guardrails instance.
   copy

   $ curl -sf http://localhost:8000/v1/health && echo 'server ready'

   • Adds RailOutcome — an engine-neutral allow, block, or transform contract returned by rail actions and enforced by both LLMRails and IORails.
   • Adds typed rail manifests so built-in rails can declare their configuration, actions, execution surfaces, requirements, and privacy properties for automatic discovery by both engines.
   • Adds output-rail checking mode to the /v1/checks endpoint, which now selects a server-loaded configuration via config_id or the server default.
   • Adds /v1/health and /healthz server health-check endpoints.
   • Adds multiple namespaced self-check rails with per-rail namespaced task prompts, enabling more than one self-check rail to run in the same configuration.

   +6 more show less

   • Adds F5 Guardrails integration to the built-in rail library.
   • Adds a canonical outbound HTTP client with connection pooling, retries, TLS, error handling, lifecycle management, and privacy-safe tracing and metrics for all built-in integrations.
   • Adds shared model telemetry and an instrumented model decorator for LLMLLMA large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. call observability.
   • Expands IORails to execute 59 of the library's 67 action-backed input and output surfaces, including community integrations and content-transforming rails, sharing implementations with LLMRails.
   • Adds message checks to IORails.
   • Non-streaming generate() and generate_async() calls on IORails now return GenerationResponse when generation options are supplied.

   11 details in Everything

   └──▷ BREAKING ON UPGRADE

   • !Custom actions using @action(output_mapping=...) must remove that argument and return an explicit RailOutcome instead.
   • !The hf-classifier install extra is removed; install transformers and torch directly to use the local classifier backend.
   • !IORails calls that supply generation options now return GenerationResponse; message lists must be passed with messages= as a keyword argument rather than positionally.
   • !Custom Colang 1 flows using the former space-separated Cleanlab, Fiddler, or GCP action names must switch to their snake_case names.
   • !/v1/checks no longer accepts inline configuration; select a server-loaded configuration via config_id or rely on the server default.
   • !Chat Completions requests must use supported role-specific OpenAI message shapes; internal event payloads, unexpected fields, and audio requests are now rejected.
2. v0.23.0 Jul 1, 2026 · issue 020

   NeMo Guardrails v0.23.0 adds tool-call rails, HF classifier rails, PIIPIIPersonally Identifiable Information — data that can identify a specific individual, such as names, email addresses, or SSNs. Cyber tools flag or redact it to limit exposure and meet privacy regulations. masking, context bloat detection, and a /v1/checks endpoint.

   └──▷ GET THIS VERSION

   copy

   $ git clone --branch v0.23.0 https://github.com/NVIDIA-NeMo/Guardrails.git
   # already have the repo? check out this version:
   $ git checkout v0.23.0

   └──▷ USE IT
   Apply Polygraf PII detection and masking to user input before it reaches the LLMLLMA large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code..
   yaml copy

   rails:
     input:
       flows:
         - polygraf pii detection
         - polygraf pii masking

   • Adds streaming and non-streaming tool callingtool callingA model's ability to answer with a structured request to run a named function or API, which the application executes and returns. It is how agents take actions. to IORails, with local rails that validate model-emitted tool calls and application-returned tool results.
   • New /v1/checks endpoint on the OpenAI-compatible server runs input or output rails standalone, returning passed, modified, or blocked status without generating a model response.
   • Adds lightweight Hugging Face classifier rails for input, output, and retrieval, backed by local Transformers, vLLM, KServe, and FMS.
   • Adds context bloat detection rails to flag oversized, repetitive, low-entropy, or padded input and retrieved content.
   • Adds Polygraf integration for PII detection and masking across input, output, and retrieval rails.

   +4 more show less

   • Replaces Annoy with exact NumPy search as the default embeddingembeddingLists of numbers a model produces to represent the meaning of text, images or code, so similar items sit close together. They power semantic search and RAG. index, removing the native C++ dependency while preserving similarity-threshold semantics.
   • Extends IORails OpenTelemetryOpenTelemetryAn open standard and set of tools for collecting traces, metrics and logs from applications. It lets services export observability data without tying instrumentation to one monitoring vendor. support with opt-in content capture and richer request, response, and token-usage span attributes.
   • LangChain integration adds support for the OpenAI Responses API and Harmony response format models.
   • Distribution wheel is now approximately ten times smaller by excluding examples and development files.

   9 details in Everything

   └──▷ BREAKING ON UPGRADE

   • !Pydantic >=2.5,<3.0 is now required; environments pinned to Pydantic 1.x must upgrade before installing v0.23.0.
3. v0.22.0 May 22, 2026 · release notes

   IORails adds streaming, OpenTelemetryOpenTelemetryAn open standard and set of tools for collecting traces, metrics and logs from applications. It lets services export observability data without tying instrumentation to one monitoring vendor., and speculative generation; LangChain becomes optional with NEMOGUARDRAILS_LLM_FRAMEWORK.

   └──▷ GET THIS VERSION

   copy

   $ git clone --branch v0.22.0 https://github.com/NVIDIA-NeMo/Guardrails.git
   # already have the repo? check out this version:
   $ git checkout v0.22.0

   └──▷ TRY IT
   Opt out of telemetry by setting the environment variable in a shell or Dockerfile.
   copy

   $ export NEMO_GUARDRAILS_NO_USAGE_STATS=1

   Use LangChain as the LLMLLMA large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. framework by setting the environment variable at runtime.
   copy

   $ export NEMOGUARDRAILS_LLM_FRAMEWORK=langchain

   Access the new testing utilities directly in Python code.
   python copy

   from nemoguardrails.testing import *
   # Use available test helpers here

   • NEMO_GUARDRAILS_NO_USAGE_STATS environment variable to disable anonymous telemetry collection
   • .config/nemoguardrails/do_not_track file to opt out of usage reporting without setting an environment variable
   • NEMOGUARDRAILS_LLM_FRAMEWORK=langchain environment variable to re-enable LangChain when using non-OpenAI providers
   • nemoguardrails.testing public testing surface for library-level test utilities
   • Speculative generation in IORails that runs input rails in parallel with the main LLM response to hide latency

   +2 more show less

   • IORails streaming output rail support for real-time guardrail evaluation on LLM outputs
   • IORails OpenTelemetry tracing and token-level metrics integration

   7 details in Everything

   └──▷ BREAKING ON UPGRADE

   • !LangChain and LangChain-providers are removed from core dependencies; they must be installed separately when NEMOGUARDRAILS_LLM_FRAMEWORK=langchain is set
   • !AsyncWorkQueue is moved from the Guardrails module to IORails, breaking existing imports from the previous location
4. v0.21.0 Mar 12, 2026 · release notes

   New IORails engine with parallel rail execution, check_async validation method, OpenAI-compatible server API, and LangChain middleware

   └──▷ GET THIS VERSION

   copy

   $ git clone --branch v0.21.0 https://github.com/NVIDIA-NeMo/Guardrails.git
   # already have the repo? check out this version:
   $ git checkout v0.21.0

   • check_async method added to LLMRails for standalone input/output rails validation without a full conversation flow
   • IORails, a new optimized Input/Output rail engine supporting parallel execution of content-safety, topic-safety, and jailbreak detection rails with logging and unique request IDs
   • Guardrails server now fully OpenAI-compatible, including the new v1/models endpoint
   • GuardrailsMiddleware added for seamless LangChain agent integration
   • CrowdStrike AIDR community integration for rail-based moderation

   +1 more show less

   • PolicyAI integration for content moderation

   6 details in Everything

   └──▷ BREAKING ON UPGRADE

   • !stream_usage and streaming metadata capture behavior changed; existing code relying on the old stream metadata structure may break
5. v0.20.0 Jan 22, 2026 · release notes

   Adds reasoning guardrail connector and GLiNER for open PIIPIIPersonally Identifiable Information — data that can identify a specific individual, such as names, email addresses, or SSNs. Cyber tools flag or redact it to limit exposure and meet privacy regulations. detection

   └──▷ GET THIS VERSION

   copy

   $ git clone --branch v0.20.0 https://github.com/NVIDIA-NeMo/Guardrails.git
   # already have the repo? check out this version:
   $ git checkout v0.20.0

   • Add support to auto select multilingual refusal bot messages
   • Adding GLiNER for PII detection (open alternative to PrivateAI)

   2 details in Everything

   └──▷ BREAKING ON UPGRADE

   • !The streaming field has been removed from the config
6. v0.19.0 Dec 3, 2025 · release notes

   Adds LangChain 1.x support with content blocks API and benchmarking scripts

   └──▷ GET THIS VERSION

   copy

   $ git clone --branch v0.19.0 https://github.com/NVIDIA-NeMo/Guardrails.git
   # already have the repo? check out this version:
   $ git checkout v0.19.0

   • langchain v1 compatibility is now supported in NeMo Guardrails.
   • Adds llm provider support for LangChain 1.x content blocks to handle reasoning traces and tool calls.
   • Adds a Procfile to run Guardrails with mock LLMsLLMsA large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. for benchmarking.
   • Adds an AIPerf run script for benchmarking workflows.

   4 details in Everything

7. v0.18.0 Nov 6, 2025 · release notes

   New BotThinking events, embeddingembeddingLists of numbers a model produces to represent the meaning of text, images or code, so similar items sit close together. They power semantic search and RAG. providers (Azure, Cohere, Google), Cisco AI Defense integration, and in-memory caching for content models.

   └──▷ GET THIS VERSION

   copy

   $ git clone --branch v0.18.0 https://github.com/NVIDIA-NeMo/Guardrails.git
   # already have the repo? check out this version:
   $ git checkout v0.18.0

   • Introduces BotThinking events to apply guardrails on an LLMLLMA large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code.'s reasoning traces
   • Adds Azure OpenAI as a supported embedding provider
   • Adds Cohere embedding integration
   • Adds Google embedding integration
   • Adds Cisco AI Defense integration

   +3 more show less

   • Adds in-memory LFU caching for content-safety, topic-control, and jailbreak detection models to reduce latency
   • Adds automatic provider inference for LangChain LLMsLLMsA large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code.
   • Adds custom HTTP header support for the ChatNVIDIA provider

   8 details in Everything

   └──▷ BREAKING ON UPGRADE

   • !Reasoning traces are extracted to a separate reasoning field instead of being prepended; use LangChain's additional_kwargs structure.
   • !stream_async now raises an error if output rail streaming is disabled.
8. v0.17.0 Oct 9, 2025 · release notes

   NeMo Guardrails v0.17.0 adds LangGraph integration, tool rails, and Trend Micro Vision One AI integration

   └──▷ GET THIS VERSION

   copy

   $ git clone --branch v0.17.0 https://github.com/NVIDIA-NeMo/Guardrails.git
   # already have the repo? check out this version:
   $ git checkout v0.17.0

   • LLMRails now supports tool call passthrough for building safe multi-agent workflows with LangGraph.
   • RunnableRails has been rewritten to support the full LangChain Runnable protocol (sync, async, streaming, batch).
   • New tool-rails subsystem provides tool input rails for message validation and processing.
   • New tool-rails subsystem provides tool output rails and validation logic.
   • Community integration added for Trend Micro Vision One AI Scanner (AI Guard) to detect harmful content, data leakage, and prompt injections.

   5 details in Everything

9. v0.16.0 Sep 5, 2025 · release notes

   Adds Pangea AI Guard and GuardrailsAI community integrations, OpenTelemetryOpenTelemetryAn open standard and set of tools for collecting traces, metrics and logs from applications. It lets services export observability data without tying instrumentation to one monitoring vendor. tracing, and LLMRails method chaining.

   └──▷ GET THIS VERSION

   copy

   $ git clone --branch v0.16.0 https://github.com/NVIDIA-NeMo/Guardrails.git
   # already have the repo? check out this version:
   $ git checkout v0.16.0

   • Adds Pangea AI Guard as a supported community integration for guardrail validation
   • Adds GuardrailsAI as a supported community integration for guardrail validation
   • Enhances the tracing system to use OpenTelemetry semantic conventions for GenAI applications
   • Allows method chaining on LLMRails instances by returning self from register_* methods

   4 details in Everything

10. v0.15.0 Aug 8, 2025 · release notes

    NeMo Guardrails v0.15.0 adds parallel rails execution, OpenTelemetryOpenTelemetryAn open standard and set of tools for collecting traces, metrics and logs from applications. It lets services export observability data without tying instrumentation to one monitoring vendor. tracing API, and external async token generators.

    └──▷ GET THIS VERSION

    copy

    $ git clone --branch v0.15.0 https://github.com/NVIDIA-NeMo/Guardrails.git
    # already have the repo? check out this version:
    $ git checkout v0.15.0

    • Update tracing to use otel api
    • Support parallel rails execution
    • Implement parallel streaming output rails execution
    • Support external async token generators

    4 details in Everything

    └──▷ BREAKING ON UPGRADE

    • !The tracing subsystem now uses the otel api instead of its previous implementation, requiring migration for existing tracing setups.
11. v0.14.1 Jul 8, 2025 · release notes

    Adds direct API key configuration support for jailbreak detection

    └──▷ GET THIS VERSION

    copy

    $ git clone --branch v0.14.1 https://github.com/NVIDIA-NeMo/Guardrails.git
    # already have the repo? check out this version:
    $ git checkout v0.14.1

12. v0.14.0 Jun 11, 2025 · release notes

    New YARA-based output guardrail for code injection detection and Clavata community integration

    └──▷ GET THIS VERSION

    copy

    $ git clone --branch v0.14.0 https://github.com/NVIDIA-NeMo/Guardrails.git
    # already have the repo? check out this version:
    $ git checkout v0.14.0

    • Adds fuzzy find provider capability to the CLI (nemo-guardrails providers fuzzy-find <query>) allowing practitioners to locate available guardrail providers via shell command.
    • Adds inline YARA rules support in the output guardrail configuration (setting type: yara_inline and providing rule: <yaml_string>), enabling direct code-injection detection without external files.
    • Adds a new community integration for Clavata LLMLLMA large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. content moderation (provider type clavata) accessible via library configuration or CLI provider selection.
    • Adds support for preserving and applying guardrails to reasoning traces, controlled by the config key preserve_reasoning_traces (boolean) in Rail YAML definitions.

    4 details in Everything

    └──▷ BREAKING ON UPGRADE

    • !The config key remove_thinking_traces is renamed to remove_reasoning_traces; existing YAML files using the old name will fail validation or ignore the setting.
    • !The deprecated argument return_context is removed; code passing this argument to library functions will raise an error.
13. v0.13.0 Mar 25, 2025 · release notes

    NeMo Guardrails v0.13.0 adds multimodalmultimodalAble to work with more than one type of input or output, such as text, images and audio. A multimodal model can combine information across those formats. rails, Fiddler integration, and reasoning trace support

    └──▷ GET THIS VERSION

    copy

    $ git clone --branch v0.13.0 https://github.com/NVIDIA-NeMo/Guardrails.git
    # already have the repo? check out this version:
    $ git checkout v0.13.0

    • FiddlerGuardrail integration enables low-latency hosted guardrail models for content safety and hallucination detection.
    • Multimodal input and output rails allow checking safety of image prompts and vision-language modelvision-language modelA vision-language model that works with images and text together, such as answering questions about a picture or describing what it contains. outputs.
    • New NemoGuardJailbreakDetect NIM support for jailbreak detection capabilities.
    • Generation metadata is now included in streaming chunks via the metadata field on chunk objects.

    4 details in Everything

14. v0.12.0 Feb 26, 2025 · release notes

    NeMo Guardrails v0.12.0 adds output rails streaming, Prompt Security integration, and score thresholds to AnalyzerEngine

    └──▷ GET THIS VERSION

    copy

    $ git clone --branch v0.12.0 https://github.com/NVIDIA-NeMo/Guardrails.git
    # already have the repo? check out this version:
    $ git checkout v0.12.0

    • Support Output Rails Streaming
    • Add pii masking capability to PrivateAI integration
    • Add embedding_params to BasicEmbeddingsIndex
    • Add score threshold to AnalyzerEngine

    4 details in Everything

15. v0.11.1 Jan 16, 2025 · release notes

    NeMo Guardrails v0.11.1 adds ContentSafety, TopicControl, and JailbreakDetect NIM connectors

    └──▷ GET THIS VERSION

    copy

    $ git clone --branch v0.11.1 https://github.com/NVIDIA-NeMo/Guardrails.git
    # already have the repo? check out this version:
    $ git checkout v0.11.1

    • ContentSafety connector: new integration for the NVIDIA NemoGuard content safety NIM microservice to detect biased or harmful outputs.
    • TopicControl connector: new integration for the NVIDIA NemoGuard topic control NIM microservice to keep conversations on approved topics.
    • JailbreakDetect connector: new integration for the NVIDIA NemoGuard jailbreak detection NIM microservice to identify jailbreak attempts.

    3 details in Everything

16. v0.11.0 Nov 19, 2024 · release notes

    New .railsignore file support, Private AI and Patronus Evaluate integrations, and Colang 2 passthrough.

    └──▷ GET THIS VERSION

    copy

    $ git clone --branch v0.11.0 https://github.com/NVIDIA-NeMo/Guardrails.git
    # already have the repo? check out this version:
    $ git checkout v0.11.0

    • Adds .railsignore file support to exclude specific files from RailsConfig rules
    • Adds Private AI Integration for private AI use cases
    • Adds Patronus Evaluate API Integration
    • Adds passthrough support in Colang 2
    • Adds sandboxed environment in Jinja2

    5 details in Everything

    └──▷ BREAKING ON UPGRADE

    • !Dropped support for Python 3.8
17. v0.10.0 Sep 27, 2024 · release notes

    New Content Safety Module, Colang 2 Migration Tool, LLMRails serialization, and Cleanlab score in NeMo Guardrails v0.10.0

    └──▷ GET THIS VERSION

    copy

    $ git clone --branch v0.10.0 https://github.com/NVIDIA-NeMo/Guardrails.git
    # already have the repo? check out this version:
    $ git checkout v0.10.0

    • Adds a ContentSafetyModule to integrate content safety models
    • Introduces a MigrationTool for converting Colang 1.0 or 2-alpha configs to Colang 2.x
    • Enables Rails Exception Handling in Colang 1.0 and 2.x
    • Adds relevance_chunks support to Colang 2 prompts
    • Implements serialization support for the LLMRails class via llmrails

    +2 more show less

    • Integrates Cleanlab's Trustworthiness Score into the library
    • Adds GCP moderation support

    7 details in Everything

18. v0.9.1 Jul 25, 2024 · release notes

    Adds --default-config-id CLI flag, custom embedding modelembedding modelLists of numbers a model produces to represent the meaning of text, images or code, so similar items sit close together. They power semantic search and RAG. support, and Got It AI / AutoAlign / Patronus Lynx integrations

    └──▷ GET THIS VERSION

    copy

    $ git clone --branch v0.9.1 https://github.com/NVIDIA-NeMo/Guardrails.git
    # already have the repo? check out this version:
    $ git checkout v0.9.1

    • Adds --default-config-id argument to the server CLI for specifying a default config
    • Enables registering and using custom embedding modelsembedding modelsLists of numbers a model produces to represent the meaning of text, images or code, so similar items sit close together. They power semantic search and RAG. via new library APIs
    • Adds Got It AI truthchecking integration for RAGRAGRetrieval-augmented generation: looking up relevant documents and adding them to the prompt before the model answers. It grounds responses in private or current data without retraining. applications
    • Adds AutoAlign guardrail integration
    • Adds Patronus Lynx integration

    5 details in Everything

19. v0.9.0 May 10, 2024 · release notes

    Adds Colang 2.0 language and runtime with parallel flows, the ... generation operator, and async actions.

    └──▷ GET THIS VERSION

    copy

    $ git clone --branch v0.9.0 https://github.com/NVIDIA-NeMo/Guardrails.git
    # already have the repo? check out this version:
    $ git checkout v0.9.0

    • _generation operator (...) for constructing output sequences within Colang 2.0 flows
    • Parallel flow execution engine supporting multiple simultaneous concurrent flows
    • Asynchronous action execution model allowing non-blocking action invocation in Colang 2.0

    3 details in Everything

    └──▷ BREAKING ON UPGRADE

    • !Colang 1.0 syntax and semantics are no longer the default; Colang 2.0 (Beta) is now included, requiring migration for existing configurations.
20. v0.8.2 Apr 1, 2024 · release notes

    Adds Vertex AI and NVIDIA AI Endpoints integrations; exposes verbose arg in RunnableRails.

    └──▷ GET THIS VERSION

    copy

    $ git clone --branch v0.8.2 https://github.com/NVIDIA-NeMo/Guardrails.git
    # already have the repo? check out this version:
    $ git checkout v0.8.2

    • Add support for Vertex AI Models into Guardrails
    • Add support for NVIDIA AI Endpoints
    • Expose the verbose parameter through RunnableRails

    3 details in Everything

21. v0.8.1 Mar 15, 2024 · release notes

    Adds streaming example for custom actions in Colang 2.0

    └──▷ GET THIS VERSION

    copy

    $ git clone --branch v0.8.1 https://github.com/NVIDIA-NeMo/Guardrails.git
    # already have the repo? check out this version:
    $ git checkout v0.8.1

22. v0.8.0 Feb 28, 2024 · release notes

    NeMo Guardrails v0.8.0 adds jailbreak heuristics, generation options, multi-config API calls, and Colang 2.0 preview

    └──▷ GET THIS VERSION

    copy

    $ git clone --branch v0.8.0 https://github.com/NVIDIA-NeMo/Guardrails.git
    # already have the repo? check out this version:
    $ git checkout v0.8.0

    • Adds a new input rail type based on jailbreak heuristics for detecting adversarial prompts.
    • Adds generation_options to control which rails trigger, what data is returned, and logging details in the response.
    • Adds support for making API calls to the guardrails server using multiple configuration ids.
    • Adds configurable stop tokens per task template.
    • Adds a preview implementation of Colang 2.0 for early testing.

    5 details in Everything

23. v0.7.0 Jan 31, 2024 · release notes

    NeMo Guardrails v0.7.0 adds Llama Guard moderation, server-side threads, and RunnableRails LangChain integration

    └──▷ GET THIS VERSION

    copy

    $ git clone --branch v0.7.0 https://github.com/NVIDIA-NeMo/Guardrails.git
    # already have the repo? check out this version:
    $ git checkout v0.7.0

    └──▷ USE IT
    Integrate NeMo Guardrails directly into a LangChain RunnableSequence or LCEL pipeline using the new wrapper.
    python copy

    guardrail_chain = RunnableRails(config_path).with_config(run_name="nemo_guardrails")

    • RunnableRails class for improved LangChain integration allowing direct use within LangChain chains.
    • Support for server-side threads in the NeMo Guardrails library API.
    • Native support for Llama Guard input and output content moderation as a new guardrail type.

    3 details in Everything

24. v0.6.1 Dec 20, 2023 · release notes

    Adds --version CLI flag to NeMo Guardrails

    └──▷ GET THIS VERSION

    copy

    $ git clone --branch v0.6.1 https://github.com/NVIDIA-NeMo/Guardrails.git
    # already have the repo? check out this version:
    $ git checkout v0.6.1

    └──▷ TRY IT
    Check the installed version of NeMo Guardrails from the terminal.
    copy

    $ nemo guardrails --version

    • Support for --version flag in the CLI

    1 detail in Everything

25. v0.6.0 Dec 13, 2023 · release notes

    NeMo Guardrails v0.6.0 adds rails configs, AlignScore fact-checking, streaming, and Presidio detection

    └──▷ GET THIS VERSION

    copy

    $ git clone --branch v0.6.0 https://github.com/NVIDIA-NeMo/Guardrails.git
    # already have the repo? check out this version:
    $ git checkout v0.6.0

    └──▷ USE IT
    Get a textual explanation of why the guardrails triggered for a given interaction.
    python copy

    explanation = llm_rails.explain()

    Load rails configuration from a Python dictionary instead of a file at runtime.
    python copy

    rails_config = RailsConfig.from_content(config=my_dict)

    • Adds rails.dialog.single_call.enabled config to reduce latency by combining guardrails and response generation in a single LLMLLMA large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. call
    • Integrates ActiveFence ActiveScore API as an input rail for the first time
    • Adds Presidio integration for sensitive data detection guardrails
    • Adds LLMRails.explain() method to get explanation of guardrail behavior
    • Supports AlignScore for fast fact-checking of LLM responses

    +4 more show less

    • Adds support for streaming LLM responses when no output rails are configured
    • Adds prompting_mode config key to control prompting modes
    • Enables loading configuration from a dictionary using RailsConfig.from_content(config=...)
    • Adds Dockerfile for building a container image

    9 details in Everything