Find why your document-backed AI answers fail.
Evaluates retrieval and answer quality in retrieval-augmented generation pipelines to identify failures and guide improvements.
When to use it
Use when evaluating retrieval-augmented generation, measuring retrieval quality, assessing answer faithfulness or relevance, generating synthetic test questions, or optimizing chunking.
Give it pipeline traces and evaluation data; it assesses retrieval and generation separately and identifies where improvements are needed.
What you provide
No setup requirements listed.
No additional actions listed.