{"rank":null,"reference":false,"run_id":"a988457822dd","task":"clinicalqa@0","model":{"id":"alibaba/qwen3.8-max-2.4t-a95b","slug":"alibaba/qwen3.8-max-2.4t-a95b","name":"Qwen3.8 Max 2.4T-A95B","type":"llm","developer":"Alibaba"},"tier":"baseline","metric":{"name":"accuracy","value":0.9062200956937799,"ci_low":0.8889952153110048,"ci_high":0.9234449760765551},"n":1045,"parse_rate":0.9942583732057416,"cost_usd":0.0,"eval_date":"2026-10-29","url":"https://medbench.kr/r/a988457822dd","status":"done","listed":true,"current_spec":true,"metrics":{"accuracy":{"value":0.9062200956937799,"ci_low":0.8889952153110048,"ci_high":0.9234449760765551,"n":1045,"n_unparsed":6},"parse_rate":0.9942583732057416,"error_rate":0.0,"by_category":{"Allergy and Immunology":{"value":0.9166666666666666,"n":36},"Cardiology":{"value":0.9325842696629213,"n":89},"Dermatology":{"value":0.9166666666666666,"n":12},"Emergency Medicine":{"value":1.0,"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":0.9285714285714286,"n":112},"General Surgery":{"value":0.896551724137931,"n":58},"Geriatric Medicine":{"value":1.0,"n":1},"Gynecology":{"value":0.7857142857142857,"n":70},"Hematology and Oncology":{"value":0.9791666666666666,"n":48},"Infectious Diseases":{"value":0.8653846153846154,"n":52},"Musculoskeletal (e.g., costochondritis)":{"value":1.0,"n":1},"Nephrology":{"value":0.8648648648648649,"n":111},"Neurology / Neurosurgery":{"value":0.9795918367346939,"n":49},"Obstetrics":{"value":0.7777777777777778,"n":18},"Ophthalmology":{"value":1.0,"n":1},"Orthopedic Surgery":{"value":0.9583333333333334,"n":24},"Otolaryngology":{"value":1.0,"n":14},"Otolaryngology (ENT)":{"value":1.0,"n":14},"Pediatrics":{"value":0.875,"n":56},"Pharmacology":{"value":1.0,"n":1},"Psychiatry":{"value":0.8888888888888888,"n":45},"Pulmonology (Respiratory Medicine)":{"value":0.9384615384615385,"n":65},"Radiology":{"value":1.0,"n":1},"Rheumatology":{"value":0.9387755102040817,"n":49},"Urology":{"value":0.8461538461538461,"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":"e96c82efe71cc3c93cb93f139bb0c91cd801d3d775eb450d4dd40a242110a721","harness_version":"p0.1","task_spec_hash":"77e37cd0ca6bded7fba03792ec1922342b3f6a7dc3ce602881ce196e6aa7377c","model_id":"alibaba/qwen3.8-max-2.4t-a95b","tokens_in":419053,"tokens_out":1840391,"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/a988457822dd/predictions.parquet","bibtex":"@misc{medbench_a988457822dd,\n  title = {Qwen3.8 Max 2.4T-A95B on Korean clinical knowledge (KMLE-level MCQ) (clinicalqa@0)},\n  author = {{medbench.kr}},\n  year = {2026},\n  howpublished = {\\url{https://medbench.kr/r/a988457822dd}},\n  note = {Evaluated 2026-10-29, harness p0.1, run key e96c82efe71c}\n}"}