API Overview

Modules

• nemoguardrails.context

• nemoguardrails.embeddings.basic

• nemoguardrails.embeddings.index

• nemoguardrails.rails.llm.config: Module for the configuration of rails.

• nemoguardrails.rails.llm.llmrails: LLM Rails entry point.

• nemoguardrails.streaming

Classes

• basic.BasicEmbeddingsIndex: Basic implementation of an embeddings index.

• basic.OpenAIEmbeddingModel: Embedding model using OpenAI API.

• basic.SentenceTransformerEmbeddingModel: Embedding model using sentence-transformers.

• index.EmbeddingModel: The embedding model is responsible for creating the embeddings.

• index.EmbeddingsIndex: The embeddings index is responsible for computing and searching a set of embeddings.

• index.IndexItem: IndexItem(text: str, meta: Dict = )

• config.CoreConfig: Settings for core internal mechanics.

• config.DialogRails: Configuration of topical rails.

• config.Document: Configuration for documents that should be used for question answering.

• config.EmbeddingSearchProvider: Configuration of a embedding search provider.

• config.FactCheckingRailConfig: Configuration data for the fact-checking rail.

• config.InputRails: Configuration of input rails.

• config.Instruction: Configuration for instructions in natural language that should be passed to the LLM.

• config.KnowledgeBaseConfig

• config.MessageTemplate: Template for a message structure.

• config.Model: Configuration of a model used by the rails engine.

• config.OutputRails: Configuration of output rails.

• config.Rails: Configuration of specific rails.

• config.RailsConfig: Configuration object for the models and the rails.

• config.RailsConfigData: Configuration data for specific rails that are supported out-of-the-box.

• config.RetrievalRails: Configuration of retrieval rails.

• config.SensitiveDataDetection: Configuration of what sensitive data should be detected.

• config.SensitiveDataDetectionOptions

• config.SingleCallConfig: Configuration for the single LLM call option for topical rails.

• config.TaskPrompt: Configuration for prompts that will be used for a specific task.

• config.UserMessagesConfig: Configuration for how the user messages are interpreted.

• llmrails.LLMRails: Rails based on a given configuration.

• streaming.StreamingHandler: Streaming async handler.

Functions

• basic.init_embedding_model: Initialize the embedding model.