{"rank":null,"reference":false,"run_id":"75310a96c0a0","task":"clinicalqa@0","model":{"id":"openai/gpt-oss-20b","slug":"openai/gpt-oss-20b","name":"gpt-oss 20B","type":"llm","developer":"OpenAI"},"tier":"baseline","metric":{"name":"accuracy","value":0.8516746411483254,"ci_low":0.830622009569378,"ci_high":0.8736842105263158},"n":1045,"parse_rate":0.9952153110047847,"cost_usd":0.09091615540000002,"eval_date":"2026-10-29","url":"https://medbench.kr/r/75310a96c0a0","status":"done","listed":true,"current_spec":true,"metrics":{"accuracy":{"value":0.8516746411483254,"ci_low":0.830622009569378,"ci_high":0.8736842105263158,"n":1045,"n_unparsed":5},"parse_rate":0.9952153110047847,"error_rate":0.0,"by_category":{"Allergy and Immunology":{"value":0.8055555555555556,"n":36},"Cardiology":{"value":0.898876404494382,"n":89},"Dermatology":{"value":0.5833333333333334,"n":12},"Emergency Medicine":{"value":0.96875,"n":32},"Endocrinology":{"value":0.8382352941176471,"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":0.8103448275862069,"n":58},"Geriatric Medicine":{"value":1.0,"n":1},"Gynecology":{"value":0.7428571428571429,"n":70},"Hematology and Oncology":{"value":0.9791666666666666,"n":48},"Infectious Diseases":{"value":0.7884615384615384,"n":52},"Musculoskeletal (e.g., costochondritis)":{"value":1.0,"n":1},"Nephrology":{"value":0.8648648648648649,"n":111},"Neurology / Neurosurgery":{"value":0.9387755102040817,"n":49},"Obstetrics":{"value":0.8333333333333334,"n":18},"Ophthalmology":{"value":1.0,"n":1},"Orthopedic Surgery":{"value":0.875,"n":24},"Otolaryngology":{"value":1.0,"n":14},"Otolaryngology (ENT)":{"value":0.7857142857142857,"n":14},"Pediatrics":{"value":0.8214285714285714,"n":56},"Pharmacology":{"value":1.0,"n":1},"Psychiatry":{"value":0.8,"n":45},"Pulmonology (Respiratory Medicine)":{"value":0.8461538461538461,"n":65},"Radiology":{"value":0.0,"n":1},"Rheumatology":{"value":0.8571428571428571,"n":49},"Urology":{"value":0.6923076923076923,"n":13},"Vascular Surgery":{"value":1.0,"n":2},"nan":{"value":0.0,"n":1}}},"config":{"max_tokens":16384,"prompt":"clinicalqa-p0","temperature":0},"run_key":"1deaf21dd56a0e6afc33e449e7f7691c4b58949d3c68a4323e1036d5f6bf112a","harness_version":"p0.1","task_spec_hash":"77e37cd0ca6bded7fba03792ec1922342b3f6a7dc3ce602881ce196e6aa7377c","model_id":"openai/gpt-oss-20b","tokens_in":460978,"tokens_out":552925,"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/75310a96c0a0/predictions.parquet","bibtex":"@misc{medbench_75310a96c0a0,\n  title = {gpt-oss 20B on Korean clinical knowledge (KMLE-level MCQ) (clinicalqa@0)},\n  author = {{medbench.kr}},\n  year = {2026},\n  howpublished = {\\url{https://medbench.kr/r/75310a96c0a0}},\n  note = {Evaluated 2026-10-29, harness p0.1, run key 1deaf21dd56a}\n}"}