Data as of Jul 25, 2026 · Based on 2,718,867 AI responses across 9,511 prompts · See how Parse measures this
SCARFgithub.com · Low evidenceSCARF is a modular evaluation framework for systematic benchmarking of Retrieval Augmented Generation (RAG) applications. It provides an end-to-end, black-box methodology to assess factual accuracy, contextual relevance, and response coherence across diverse RAG frameworks.
Parse Score