RAG Evaluation
Connect your RAG pipeline or knowledge base via HTTP, run structured test scenarios, evaluate retrieval context relevance, faithfulness, and answer groundedness, diagnose root causes, and automatically build regression test suites from failures.
RAG Pipeline Evaluation Standards (RAG Triad)
Ensure high retrieval accuracy, grounded responses, and zero unverified factual hallucinations.
Context Relevance
Measures whether vector search and hybrid rerankers retrieved solely relevant chunks without injecting noisy, distracting passages into the prompt context.
Faithfulness & Groundedness
Verifies every single claim made in the generated answer maps directly to verifiable retrieved evidence, penalizing hallucinations.
Answer Relevance & Completeness
Calculates how directly and thoroughly the response addresses the user question without superfluous filler or omitted constraints.