Reference · Glossary
RAGAS
Last updated
An open-source evaluation library built specifically for **RAG pipelines** — it scores answers on metrics like faithfulness (is the answer grounded in retrieved text?), answer relevance, and context precision/recall, often using an LLM as the judge.
#When to use
Grading a RAG app beyond "it looks fine to me" — automated, repeatable scores you can track across prompt or chunking changes.
#When not to
Evaluating a plain chat app with no retrieval step — RAGAS's metrics are specifically about how well answers are grounded in retrieved context.
#Example
from ragas import evaluate
from ragas.metrics import faithfulness, answer_relevancy
results = evaluate(dataset, metrics=[faithfulness, answer_relevancy])
print(results)