{"rank":null,"reference":false,"run_id":"a05229db51db","task":"clinicalqa@0","model":{"id":"google/medgemma-27b","slug":"google/medgemma-27b","name":"MedGemma 27B","type":"llm","developer":"Google"},"tier":"baseline","metric":{"name":"accuracy","value":0.8746411483253589,"ci_low":0.8545454545454545,"ci_high":0.8937799043062201},"n":1045,"parse_rate":1.0,"cost_usd":0.36981200999999997,"eval_date":"2026-10-29","url":"https://medbench.kr/r/a05229db51db","status":"done","listed":true,"current_spec":true,"metrics":{"accuracy":{"value":0.8746411483253589,"ci_low":0.8545454545454545,"ci_high":0.8937799043062201,"n":1045,"n_unparsed":0},"parse_rate":1.0,"error_rate":0.0,"parser":"p0.2","by_category":{"Allergy and Immunology":{"value":0.8611111111111112,"n":36},"Cardiology":{"value":0.9438202247191011,"n":89},"Dermatology":{"value":0.9166666666666666,"n":12},"Emergency Medicine":{"value":0.96875,"n":32},"Endocrinology":{"value":0.8529411764705882,"n":68},"Endocrinology, metabolic, and nutritional disorders":{"value":1.0,"n":1},"Forensic Medicine":{"value":1.0,"n":1},"Gastroenterology":{"value":0.8928571428571429,"n":112},"General Surgery":{"value":0.9310344827586207,"n":58},"Geriatric Medicine":{"value":1.0,"n":1},"Gynecology":{"value":0.7571428571428571,"n":70},"Hematology and Oncology":{"value":0.9166666666666666,"n":48},"Infectious Diseases":{"value":0.9038461538461539,"n":52},"Musculoskeletal (e.g., costochondritis)":{"value":1.0,"n":1},"Nephrology":{"value":0.8558558558558559,"n":111},"Neurology / Neurosurgery":{"value":0.8979591836734694,"n":49},"Obstetrics":{"value":0.8888888888888888,"n":18},"Ophthalmology":{"value":1.0,"n":1},"Orthopedic Surgery":{"value":0.9166666666666666,"n":24},"Otolaryngology":{"value":1.0,"n":14},"Otolaryngology (ENT)":{"value":0.9285714285714286,"n":14},"Pediatrics":{"value":0.8392857142857143,"n":56},"Pharmacology":{"value":1.0,"n":1},"Psychiatry":{"value":0.7555555555555555,"n":45},"Pulmonology (Respiratory Medicine)":{"value":0.8461538461538461,"n":65},"Radiology":{"value":1.0,"n":1},"Rheumatology":{"value":0.7959183673469388,"n":49},"Urology":{"value":0.9230769230769231,"n":13},"Vascular Surgery":{"value":1.0,"n":2},"nan":{"value":1.0,"n":1}}},"config":{"max_tokens":8192,"prompt":"clinicalqa-p0","temperature":0},"run_key":"c4b93899fc5a80e52f5dbafd5f1b8b2c7a6ea2a2824c643186f29c0b71071b10","harness_version":"p0.1","task_spec_hash":"77e37cd0ca6bded7fba03792ec1922342b3f6a7dc3ce602881ce196e6aa7377c","model_id":"google/medgemma-27b","tokens_in":399078,"tokens_out":919779,"dataset":{"slug":"clinicalqa","title":"ClinicalQA: Korean Medical Knowledge QA Benchmark","license":"CC BY 4.0 (upstream Apache-2.0)","license_url":"https://creativecommons.org/licenses/by/4.0/","citation":"SNUH-HARI. ClinicalQA: Korean Medical Knowledge QA Benchmark. Seoul National University Hospital Healthcare AI Research Institute, 2025. doi:10.23421/2026.0800"},"predictions_url":"https://medbench.kr/artifacts/a05229db51db/predictions.parquet","bibtex":"@misc{medbench_a05229db51db,\n  title = {MedGemma 27B on Korean clinical knowledge (KMLE-level MCQ) (clinicalqa@0)},\n  author = {{medbench.kr}},\n  year = {2026},\n  howpublished = {\\url{https://medbench.kr/r/a05229db51db}},\n  note = {Evaluated 2026-10-29, harness p0.1, run key c4b93899fc5a}\n}"}