snuh/hari-q2.5-thinking

The snuh/hari-q2.5-thinking model is a 72.7 billion parameter Large Language Model (LLM) developed by the Healthcare AI Research Institute (HARI) at Seoul National University Hospital (SNUH). Fine-tuned on Korean medical question-answering (QA) style data, it excels in clinical reasoning and medical education. This model achieves 89.2% accuracy on the Korean Medical Licensing Examination (KMLE) and 88.36% on the USMLE QA benchmark, making it suitable for domain-specific medical inference and decision support.