{"rank":null,"reference":false,"run_id":"33d8b39d1c90","task":"clinicalqa@0","model":{"id":"openai/gpt-oss-120b","slug":"openai/gpt-oss-120b","name":"gpt-oss 120B","type":"llm","developer":"OpenAI"},"tier":"baseline","metric":{"name":"accuracy","value":0.8822966507177034,"ci_low":0.8631578947368421,"ci_high":0.9014354066985646},"n":1045,"parse_rate":1.0,"cost_usd":0.0720534069,"eval_date":"2026-10-29","url":"https://medbench.kr/r/33d8b39d1c90","status":"done","listed":true,"current_spec":true,"metrics":{"accuracy":{"value":0.8822966507177034,"ci_low":0.8631578947368421,"ci_high":0.9014354066985646,"n":1045,"n_unparsed":0},"parse_rate":1.0,"error_rate":0.0,"by_category":{"Allergy and Immunology":{"value":0.8888888888888888,"n":36},"Cardiology":{"value":0.9101123595505618,"n":89},"Dermatology":{"value":0.75,"n":12},"Emergency Medicine":{"value":0.9375,"n":32},"Endocrinology":{"value":0.8823529411764706,"n":68},"Endocrinology, metabolic, and nutritional disorders":{"value":1.0,"n":1},"Forensic Medicine":{"value":1.0,"n":1},"Gastroenterology":{"value":0.8660714285714286,"n":112},"General Surgery":{"value":0.9137931034482759,"n":58},"Geriatric Medicine":{"value":1.0,"n":1},"Gynecology":{"value":0.7857142857142857,"n":70},"Hematology and Oncology":{"value":1.0,"n":48},"Infectious Diseases":{"value":0.8269230769230769,"n":52},"Musculoskeletal (e.g., costochondritis)":{"value":1.0,"n":1},"Nephrology":{"value":0.8738738738738738,"n":111},"Neurology / Neurosurgery":{"value":0.9183673469387755,"n":49},"Obstetrics":{"value":0.7222222222222222,"n":18},"Ophthalmology":{"value":1.0,"n":1},"Orthopedic Surgery":{"value":0.9583333333333334,"n":24},"Otolaryngology":{"value":1.0,"n":14},"Otolaryngology (ENT)":{"value":0.7857142857142857,"n":14},"Pediatrics":{"value":0.8392857142857143,"n":56},"Pharmacology":{"value":0.0,"n":1},"Psychiatry":{"value":0.8666666666666667,"n":45},"Pulmonology (Respiratory Medicine)":{"value":0.9230769230769231,"n":65},"Radiology":{"value":1.0,"n":1},"Rheumatology":{"value":0.9183673469387755,"n":49},"Urology":{"value":0.8461538461538461,"n":13},"Vascular Surgery":{"value":1.0,"n":2},"nan":{"value":1.0,"n":1}}},"config":{"max_tokens":16384,"prompt":"clinicalqa-p0","temperature":0},"run_key":"baa21ccda3940875ee6892b2fc59537b3df89d9736eab50197e678134b110b0d","harness_version":"p0.1","task_spec_hash":"77e37cd0ca6bded7fba03792ec1922342b3f6a7dc3ce602881ce196e6aa7377c","model_id":"openai/gpt-oss-120b","tokens_in":459933,"tokens_out":299878,"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/33d8b39d1c90/predictions.parquet","bibtex":"@misc{medbench_33d8b39d1c90,\n  title = {gpt-oss 120B on Korean clinical knowledge (KMLE-level MCQ) (clinicalqa@0)},\n  author = {{medbench.kr}},\n  year = {2026},\n  howpublished = {\\url{https://medbench.kr/r/33d8b39d1c90}},\n  note = {Evaluated 2026-10-29, harness p0.1, run key baa21ccda394}\n}"}