[{"rank":1,"reference":false,"run_id":"4c3036eee37b","task":"kor-surgical-robot-voice-command@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":"standard","metric":{"name":"exact_match","value":0.9474747474747475,"ci_low":0.9164924711757237,"ci_high":0.9741035856573705},"n":2085,"parse_rate":1.0,"cost_usd":0.0,"eval_date":"2026-10-29","url":"https://medbench.kr/r/4c3036eee37b","rank_range":null,"rank_shared":1,"rank_label":"1"},{"rank":2,"reference":false,"run_id":"da9bb98f3974","task":"kor-surgical-robot-voice-command@0","model":{"id":"deepseek/deepseek-v4.1-flash-552b-a16b","slug":"deepseek/deepseek-v4.1-flash-552b-a16b","name":"DeepSeek V4.1 Flash 552B-A16B","type":"llm","developer":"DeepSeek"},"tier":"standard","metric":{"name":"exact_match","value":0.9292929292929293,"ci_low":0.8932183266999603,"ci_high":0.9617749839417051},"n":2085,"parse_rate":1.0,"cost_usd":0.0,"eval_date":"2026-10-29","url":"https://medbench.kr/r/da9bb98f3974","rank_range":null,"rank_shared":2,"rank_label":"2"},{"rank":3,"reference":false,"run_id":"704eca6b0f7d","task":"kor-surgical-robot-voice-command@0","model":{"id":"alibaba/qwen3.8-flash-next-180b-a6b","slug":"alibaba/qwen3.8-flash-next-180b-a6b","name":"Qwen3.8 Flash Next 180B-A6B","type":"llm","developer":"Alibaba"},"tier":"standard","metric":{"name":"exact_match","value":0.9212121212121213,"ci_low":0.882011004721458,"ci_high":0.9555131677751556},"n":2085,"parse_rate":1.0,"cost_usd":0.0,"eval_date":"2026-10-29","url":"https://medbench.kr/r/704eca6b0f7d","rank_range":null,"rank_shared":3,"rank_label":"3"},{"rank":4,"reference":false,"run_id":"9bba2b5b5329","task":"kor-surgical-robot-voice-command@0","model":{"id":"google/medgemma-1.5-4b","slug":"google/medgemma-1.5-4b","name":"MedGemma 1.5 4B","type":"llm","developer":"Google"},"tier":"standard","metric":{"name":"exact_match","value":0.22828282828282828,"ci_low":0.17737069210242434,"ci_high":0.28113798283261804},"n":2085,"parse_rate":0.9985611510791367,"cost_usd":0.07845542000000001,"eval_date":"2026-10-29","url":"https://medbench.kr/r/9bba2b5b5329","rank_range":null,"rank_shared":4,"rank_label":"4"},{"rank":null,"reference":true,"run_id":"a923656cf11c","task":"kor-surgical-robot-voice-command@0","model":{"id":"medbench/prior","slug":"medbench/prior","name":"Prior (ignores the input)","type":"baseline","developer":"medbench.kr"},"tier":"baseline","metric":{"name":"exact_match","value":0.13333333333333333,"ci_low":0.08751759172666769,"ci_high":0.1916674901185771},"n":2085,"parse_rate":1.0,"cost_usd":0.0,"eval_date":"2026-10-29","url":"https://medbench.kr/r/a923656cf11c","rank_range":null,"rank_shared":null,"rank_label":"ref"}]