foundry_grader_azure_ai: Azure AI built-in evaluator grader

In foundryR: Data Frame Workflows for 'Microsoft Foundry'

Azure AI built-in evaluator grader

Description

Reference an Azure AI Foundry built-in evaluator (a ⁠builtin.*⁠ ID such as builtin.coherence or builtin.groundedness) as a grader. This grader type is only available on the project-scoped Foundry endpoint.

Usage

foundry_grader_azure_ai(
  name,
  evaluator_name,
  initialization_parameters = NULL,
  data_mapping = NULL,
  evaluator_version = NULL
)

Arguments

name	Character. Grader name shown in results.
evaluator_name	Character. The evaluator ID, e.g. "builtin.coherence".
initialization_parameters	List. Optional parameters passed to the evaluator, e.g. list(model = "gpt-5-nano") for model-graded evaluators.
data_mapping	Named list. Optional mapping from evaluator inputs to dataset templates, e.g. list(query = "{{item.query}}", response = "{{sample.output_text}}").
evaluator_version	Character. Optional evaluator version. Defaults to the latest version on the service when omitted.

Value

A named list describing an azure_ai_evaluator grader.

Examples

foundry_grader_azure_ai(
  name = "coherence",
  evaluator_name = "builtin.coherence",
  initialization_parameters = list(model = "gpt-5-nano"),
  data_mapping = list(
    query = "{{item.query}}",
    response = "{{sample.output_text}}"
  )
)

foundryR documentation built on Sept. 25, 2026, 1:10 a.m.