Data as of Jul 25, 2026 · Based on 2,718,867 AI responses across 9,511 prompts · See how Parse measures this
MedCPTgithub.com · Low evidenceMedCPT is a contrastive pre-trained transformer model for zero-shot biomedical information retrieval, trained on 255 million query-article pairs from PubMed search logs. It includes a dense retriever with separate query and article encoders, plus a cross-encoder re-ranker, all based on PubMedBERT.
Parse Score