github.com· Hosted on GitHub

What AI says about RAGEval

RAGEval is a framework designed to automatically generate evaluation datasets for assessing the knowledge usage capabilities of Large Language Models in domain-specific Retrieval Augmented Generation scenarios.

Updated Nov 14, 2026Reviewed by Dimitry Apollonsky

How we measure

Sites AI cites

50%
- rag-evaluation · PyPI (opens in a new tab) · 50%
50%
- RAG Eval (opens in a new tab) · 50%

Make RAGEval the brand AI recommends.

Give your agent the answers buyers see, the pages AI cites, and a record of what your team has already tried. It can pick the next move and check what changed after you made it.

Track RAGEval free