Source: Ragas docs: Available metrics

Documentation of the open-source Ragas library's RAG metrics (context precision/recall, faithfulness, response relevancy, noise sensitivity).

Contents
  • What: library documentation
  • Author: Ragas (Exploding Gradients)
  • Date: page dated 2025-12-09, read 2026-11-14
  • Original: List of available metrics

Summary

Catalogue of the metrics in the Ragas library; LLM-based metrics may use one or more LLM calls to produce a score.

Key claims

  • RAG metrics: Context Precision, Context Recall, Context Entities Recall, Noise Sensitivity, Response Relevancy, Faithfulness, plus multimodal variants.
  • Faithfulness = claims in the response supported by the retrieved context / total claims, in [0, 1].
  • Vectara’s HHEM-2.1-Open, a small T5-based classifier, can replace the LLM for the claim-checking step (the docs list it under the legacy metrics API, which “will be deprecated in version 0.4 and removed in version 1.0”).

Feeds

Retrieval-augmented generation

Suggest changes →