Adapter

Adapter for guardrail components.

NeMoLMAdapter(**kwargs)

Bases: BaseChatModel

LangChain ChatModel that delegates to a gllm_inference LM invoker.

Configuration parameters are expected to be provided via the NeMo Guardrails model config under the model's parameters. At minimum, you should pass:

• model: e.g. "openai/gpt-4o" or "azure-openai/gpt-4o-mini".
• credentials: The credentials for the language model.
• model_kwargs (optional): dict with keys required by our invoker builder, including any provider-specific fields (e.g. azure_endpoint, azure_deployment, base_url, etc.) and default_hyperparameters.

Initializes a new instance of the NeMoLMAdapter class.

Parameters:

Name	Type	Description	Default
**kwargs	Any	Additional configuration for the language model.	{}