{"rank":null,"reference":false,"run_id":"a4cc8ae1a524","task":"clinicalqa@0","model":{"slug":"kgpu/glm-5.3-ax-753b-a40b","name":"GLM-5.3 753B-A40B","type":"llm","provider":"kgpu"},"tier":"baseline","metric":{"name":"accuracy","value":0.9090909090909091,"ci_low":0.8918421052631579,"ci_high":0.9263157894736842},"n":1045,"parse_rate":0.9980861244019139,"cost_usd":0.0,"eval_date":"2026-10-29","url":"https://medbench.kr/r/a4cc8ae1a524","status":"done","listed":true,"current_spec":true,"metrics":{"accuracy":{"value":0.9090909090909091,"ci_low":0.8918421052631579,"ci_high":0.9263157894736842,"n":1045,"n_unparsed":2},"parse_rate":0.9980861244019139,"by_category":{"Allergy and Immunology":{"value":0.9722222222222222,"n":36},"Cardiology":{"value":0.9213483146067416,"n":89},"Dermatology":{"value":0.9166666666666666,"n":12},"Emergency Medicine":{"value":1.0,"n":32},"Endocrinology":{"value":0.9264705882352942,"n":68},"Endocrinology, metabolic, and nutritional disorders":{"value":1.0,"n":1},"Forensic Medicine":{"value":1.0,"n":1},"Gastroenterology":{"value":0.9464285714285714,"n":112},"General Surgery":{"value":0.9137931034482759,"n":58},"Geriatric Medicine":{"value":1.0,"n":1},"Gynecology":{"value":0.8,"n":70},"Hematology and Oncology":{"value":0.9791666666666666,"n":48},"Infectious Diseases":{"value":0.8846153846153846,"n":52},"Musculoskeletal (e.g., costochondritis)":{"value":1.0,"n":1},"Nephrology":{"value":0.8378378378378378,"n":111},"Neurology / Neurosurgery":{"value":0.9387755102040817,"n":49},"Obstetrics":{"value":0.8333333333333334,"n":18},"Ophthalmology":{"value":1.0,"n":1},"Orthopedic Surgery":{"value":1.0,"n":24},"Otolaryngology":{"value":1.0,"n":14},"Otolaryngology (ENT)":{"value":0.9285714285714286,"n":14},"Pediatrics":{"value":0.9107142857142857,"n":56},"Pharmacology":{"value":1.0,"n":1},"Psychiatry":{"value":0.8666666666666667,"n":45},"Pulmonology (Respiratory Medicine)":{"value":0.9076923076923077,"n":65},"Radiology":{"value":1.0,"n":1},"Rheumatology":{"value":0.8979591836734694,"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":"14b562a152460ac7214c4eabcfc174363d08845169bd465f386a4be0802ca56a","harness_version":"p0.1","task_spec_hash":"77e37cd0ca6bded7fba03792ec1922342b3f6a7dc3ce602881ce196e6aa7377c","model_digest":"kgpu/glm-5.3-ax-753b-a40b","tokens_in":478883,"tokens_out":260058,"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/a4cc8ae1a524/predictions.parquet","bibtex":"@misc{medbench_a4cc8ae1a524,\n  title = {GLM-5.3 753B-A40B on Korean clinical knowledge (KMLE-level MCQ) (clinicalqa@0)},\n  author = {{medbench.kr}},\n  year = {2026},\n  howpublished = {\\url{https://medbench.kr/r/a4cc8ae1a524}},\n  note = {Evaluated 2026-10-29, harness p0.1, run key 14b562a15246}\n}"}