Docs
...
Reciprocal rank

Reciprocal rank evaluation metric

Last updated: Mar 05, 2025
Reciprocal rank evaluation metric

The reciprocal rank metric measures the reciprocal rank of the first relevant context.

Metric details

Reciprocal rank is a retrieval quality metric for generative AI quality evaluations that measures the quality of how a retrieval system ranks relevant contexts. Retrieval quality metrics are calculated with LLM-as-a-judge models.

Scope

The reciprocal rank metric evaluates generative AI assets only.

  • Types of AI assets: Prompt templates
  • Generative AI tasks: Retrieval Augmented Generation (RAG)
  • Supported languages: English

Scores and values

The reciprocal rank metric score indicates whether relevant contexts are retrieved and where they are ranked. Higher scores indicate that the first relevant context is ranked higher. Lower scores indicate that the rist relevant context is ranked lower.

  • Range of values: 0.0-1.0
  • Best possible score: 1.0
  • Ratios:
    • At 0: None of the relevant contexts are retrieved
    • At 1: The first relevant context is at the first position.

Settings

  • Thresholds:
    • Lower bound: 0
    • Upper bound: 1

Parent topic: Evaluation metrics