{"rank":null,"reference":false,"run_id":"2224a6c5f8d6","task":"clinicalqa@0","model":{"slug":"openrouter/deepseek/deepseek-v4.1-flash-1017","name":"DeepSeek V4.1 Flash (2026-10-17)","type":"llm","provider":"openrouter"},"tier":"baseline","metric":{"name":"accuracy","value":0.9043062200956937,"ci_low":0.8870813397129187,"ci_high":0.9224880382775119},"n":1045,"parse_rate":0.9980861244019139,"cost_usd":0.47307404999999997,"eval_date":"2026-10-29","url":"https://medbench.kr/r/2224a6c5f8d6","status":"done","listed":true,"current_spec":true,"metrics":{"accuracy":{"value":0.9043062200956937,"ci_low":0.8870813397129187,"ci_high":0.9224880382775119,"n":1045,"n_unparsed":2},"parse_rate":0.9980861244019139,"error_rate":0.0,"by_category":{"Allergy and Immunology":{"value":0.9166666666666666,"n":36},"Cardiology":{"value":0.9550561797752809,"n":89},"Dermatology":{"value":0.8333333333333334,"n":12},"Emergency Medicine":{"value":1.0,"n":32},"Endocrinology":{"value":0.9117647058823529,"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":0.8285714285714286,"n":70},"Hematology and Oncology":{"value":0.9791666666666666,"n":48},"Infectious Diseases":{"value":0.8076923076923077,"n":52},"Musculoskeletal (e.g., costochondritis)":{"value":1.0,"n":1},"Nephrology":{"value":0.8738738738738738,"n":111},"Neurology / Neurosurgery":{"value":0.9591836734693877,"n":49},"Obstetrics":{"value":0.7777777777777778,"n":18},"Ophthalmology":{"value":1.0,"n":1},"Orthopedic Surgery":{"value":0.9583333333333334,"n":24},"Otolaryngology":{"value":0.9285714285714286,"n":14},"Otolaryngology (ENT)":{"value":0.7857142857142857,"n":14},"Pediatrics":{"value":0.875,"n":56},"Pharmacology":{"value":1.0,"n":1},"Psychiatry":{"value":0.9111111111111111,"n":45},"Pulmonology (Respiratory Medicine)":{"value":0.8923076923076924,"n":65},"Radiology":{"value":1.0,"n":1},"Rheumatology":{"value":0.9183673469387755,"n":49},"Urology":{"value":0.9230769230769231,"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":"4a339f14894dc4d3e0deac8f1378217632804927a48f01bd95438f10fc655610","harness_version":"p0.1","task_spec_hash":"77e37cd0ca6bded7fba03792ec1922342b3f6a7dc3ce602881ce196e6aa7377c","model_digest":"openrouter/deepseek/deepseek-v4.1-flash-1017","tokens_in":432343,"tokens_out":680371,"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/2224a6c5f8d6/predictions.parquet","bibtex":"@misc{medbench_2224a6c5f8d6,\n  title = {DeepSeek V4.1 Flash (2026-10-17) on Korean clinical knowledge (KMLE-level MCQ) (clinicalqa@0)},\n  author = {{medbench.kr}},\n  year = {2026},\n  howpublished = {\\url{https://medbench.kr/r/2224a6c5f8d6}},\n  note = {Evaluated 2026-10-29, harness p0.1, run key 4a339f14894d}\n}"}