Data as of Jul 25, 2026 · Based on 2,718,867 AI responses across 9,511 prompts · See how Parse measures this
MS MARCOmicrosoft.github.ioMS MARCO is a collection of datasets from Microsoft focused on deep learning in search, starting with a question answering dataset of 100,000 real Bing questions. It includes datasets for passage ranking, document ranking, keyphrase extraction, and conversational search, with the goal of advancing AI research.
Parse Score
