{"source":"AIMap","schema":"benchmark/v1","updated":"2026-10-22","benchmark":{"id":"cmmlu","name":"CMMLU","category":"chinese","unit":"%","higherIsBetter":true,"status":"stale","sources":["cmmlu"],"primaryVariant":"five-shot","url":"https://github.com/haonan-li/CMMLU","year":2023,"desc":{"en":"Chinese multitask knowledge benchmark: 67 subjects including China-specific topics.","zh-hans":"中文多任务知识基准：67 个学科，含中国特有主题。"},"measures":{"en":"Average accuracy (five-shot / zero-shot).","zh-hans":"平均准确率（5-shot / 0-shot）。"},"limits":{"en":"Leaderboard last updated in 2024; mostly older models; data licence is non-commercial, so only the top entries are cited.","zh-hans":"榜单最后更新于 2024 年，以旧模型为主；数据为非商业许可，本站只引用前几名。"},"contamination":"high","contaminationNote":{"en":"Public test set.","zh-hans":"测试集公开。"},"stats":{"results":20,"models":12,"latest":"2024-12-06","bySourceType":{"official":20}}},"sources":[{"id":"cmmlu","name":"CMMLU leaderboard (GitHub README)","license":"CC BY-NC-SA 4.0","policy":"limited","attribution":"Li et al., CMMLU (CC BY-NC-SA 4.0)","site":"https://github.com/haonan-li/CMMLU"}],"results":[{"model":"lingzhi-72b-chat","modelName":"Lingzhi-72B-chat","provider":null,"rawName":"Lingzhi-72B-chat","variant":"five-shot","setting":"Five-shot average","score":90.26,"unit":"%","rank":1,"rankComputedBySite":false,"date":"2024-12-06","sourceType":"official","source":"cmmlu","sourceUrl":"https://github.com/haonan-li/CMMLU#排行榜","collected":"2026-10-10"},{"model":"telechat2-35b","modelName":"Telechat2-35B","provider":null,"rawName":"Telechat2-35B","variant":"five-shot","setting":"Five-shot average","score":90.16,"unit":"%","rank":2,"rankComputedBySite":false,"date":"2024-12-06","sourceType":"official","source":"cmmlu","sourceUrl":"https://github.com/haonan-li/CMMLU#排行榜","collected":"2026-10-10"},{"model":"spark-4-0","modelName":"Spark 4.0-2024-10-14","provider":"iFlytek","rawName":"Spark 4.0-2024-10-14","variant":"five-shot","setting":"Five-shot average","score":90.07,"unit":"%","rank":3,"rankComputedBySite":false,"date":"2024-12-06","sourceType":"official","source":"cmmlu","sourceUrl":"https://github.com/haonan-li/CMMLU#排行榜","collected":"2026-10-10"},{"model":"qwen2-72b","modelName":"Qwen2-72B","provider":"Alibaba (Qwen)","rawName":"Qwen2-72B","variant":"five-shot","setting":"Five-shot average","score":89.65,"unit":"%","rank":4,"rankComputedBySite":false,"date":"2024-12-06","sourceType":"official","source":"cmmlu","sourceUrl":"https://github.com/haonan-li/CMMLU#排行榜","collected":"2026-10-10"},{"model":"jiutian","modelName":"Jiutian-大模型","provider":null,"rawName":"Jiutian-大模型","variant":"five-shot","setting":"Five-shot average","score":88.59,"unit":"%","rank":5,"rankComputedBySite":false,"date":"2024-12-06","sourceType":"official","source":"cmmlu","sourceUrl":"https://github.com/haonan-li/CMMLU#排行榜","collected":"2026-10-10"},{"model":"qwen1-5-110b","modelName":"Qwen1.5-110B","provider":"Alibaba (Qwen)","rawName":"Qwen1.5-110B","variant":"five-shot","setting":"Five-shot average","score":88.32,"unit":"%","rank":6,"rankComputedBySite":false,"date":"2024-12-06","sourceType":"official","source":"cmmlu","sourceUrl":"https://github.com/haonan-li/CMMLU#排行榜","collected":"2026-10-10"},{"model":"jiutian-57b","modelName":"JIUTIAN-57B","provider":null,"rawName":"JIUTIAN-57B","variant":"five-shot","setting":"Five-shot average","score":87.39,"unit":"%","rank":7,"rankComputedBySite":false,"date":"2024-12-06","sourceType":"official","source":"cmmlu","sourceUrl":"https://github.com/haonan-li/CMMLU#排行榜","collected":"2026-10-10"},{"model":"qwen2-5-72b","modelName":"Qwen2.5-72B","provider":"Alibaba (Qwen)","rawName":"Qwen2.5-72B","variant":"five-shot","setting":"Five-shot average","score":85.67,"unit":"%","rank":8,"rankComputedBySite":false,"date":"2024-12-06","sourceType":"official","source":"cmmlu","sourceUrl":"https://github.com/haonan-li/CMMLU#排行榜","collected":"2026-10-10"},{"model":"youyuanjian","modelName":"Youyuanjian","provider":null,"rawName":"Youyuanjian","variant":"five-shot","setting":"Five-shot average","score":83.72,"unit":"%","rank":9,"rankComputedBySite":false,"date":"2024-12-06","sourceType":"official","source":"cmmlu","sourceUrl":"https://github.com/haonan-li/CMMLU#排行榜","collected":"2026-10-10"},{"model":"qwen1-5-72b","modelName":"Qwen1.5-72B","provider":"Alibaba (Qwen)","rawName":"Qwen1.5-72B","variant":"five-shot","setting":"Five-shot average","score":83.54,"unit":"%","rank":10,"rankComputedBySite":false,"date":"2024-12-06","sourceType":"official","source":"cmmlu","sourceUrl":"https://github.com/haonan-li/CMMLU#排行榜","collected":"2026-10-10"},{"model":"spark-4-0","modelName":"Spark 4.0-2024-10-14","provider":"iFlytek","rawName":"Spark 4.0-2024-10-14","variant":"zero-shot","setting":"Zero-shot average","score":90.97,"unit":"%","rank":1,"rankComputedBySite":false,"date":"2024-12-06","sourceType":"official","source":"cmmlu","sourceUrl":"https://github.com/haonan-li/CMMLU#排行榜","collected":"2026-10-10"},{"model":"telechat2-35b","modelName":"Telechat2-35B","provider":null,"rawName":"Telechat2-35B","variant":"zero-shot","setting":"Zero-shot average","score":90.49,"unit":"%","rank":2,"rankComputedBySite":false,"date":"2024-12-06","sourceType":"official","source":"cmmlu","sourceUrl":"https://github.com/haonan-li/CMMLU#排行榜","collected":"2026-10-10"},{"model":"lingzhi-72b-chat","modelName":"Lingzhi-72B-chat","provider":null,"rawName":"Lingzhi-72B-chat","variant":"zero-shot","setting":"Zero-shot average","score":90.07,"unit":"%","rank":3,"rankComputedBySite":false,"date":"2024-12-06","sourceType":"official","source":"cmmlu","sourceUrl":"https://github.com/haonan-li/CMMLU#排行榜","collected":"2026-10-10"},{"model":"qwen1-5-110b","modelName":"Qwen1.5-110B","provider":"Alibaba (Qwen)","rawName":"Qwen1.5-110B","variant":"zero-shot","setting":"Zero-shot average","score":87.64,"unit":"%","rank":4,"rankComputedBySite":false,"date":"2024-12-06","sourceType":"official","source":"cmmlu","sourceUrl":"https://github.com/haonan-li/CMMLU#排行榜","collected":"2026-10-10"},{"model":"qwen2-72b","modelName":"Qwen2-72B","provider":"Alibaba (Qwen)","rawName":"Qwen2-72B","variant":"zero-shot","setting":"Zero-shot average","score":87.47,"unit":"%","rank":5,"rankComputedBySite":false,"date":"2024-12-06","sourceType":"official","source":"cmmlu","sourceUrl":"https://github.com/haonan-li/CMMLU#排行榜","collected":"2026-10-10"},{"model":"qwen2-5-72b","modelName":"Qwen2.5-72B","provider":"Alibaba (Qwen)","rawName":"Qwen2.5-72B","variant":"zero-shot","setting":"Zero-shot average","score":84.7,"unit":"%","rank":6,"rankComputedBySite":false,"date":"2024-12-06","sourceType":"official","source":"cmmlu","sourceUrl":"https://github.com/haonan-li/CMMLU#排行榜","collected":"2026-10-10"},{"model":"pci-transgpt","modelName":"PCI-TransGPT","provider":null,"rawName":"PCI-TransGPT","variant":"zero-shot","setting":"Zero-shot average","score":82.44,"unit":"%","rank":7,"rankComputedBySite":false,"date":"2024-12-06","sourceType":"official","source":"cmmlu","sourceUrl":"https://github.com/haonan-li/CMMLU#排行榜","collected":"2026-10-10"},{"model":"qwen1-5-72b","modelName":"Qwen1.5-72B","provider":"Alibaba (Qwen)","rawName":"Qwen1.5-72B","variant":"zero-shot","setting":"Zero-shot average","score":81.81,"unit":"%","rank":8,"rankComputedBySite":false,"date":"2024-12-06","sourceType":"official","source":"cmmlu","sourceUrl":"https://github.com/haonan-li/CMMLU#排行榜","collected":"2026-10-10"},{"model":"qwen1-5-32b","modelName":"Qwen1.5-32B","provider":"Alibaba (Qwen)","rawName":"Qwen1.5-32B","variant":"zero-shot","setting":"Zero-shot average","score":81.47,"unit":"%","rank":9,"rankComputedBySite":false,"date":"2024-12-06","sourceType":"official","source":"cmmlu","sourceUrl":"https://github.com/haonan-li/CMMLU#排行榜","collected":"2026-10-10"},{"model":"youyuanjian","modelName":"Youyuanjian","provider":null,"rawName":"Youyuanjian","variant":"zero-shot","setting":"Zero-shot average","score":81.19,"unit":"%","rank":10,"rankComputedBySite":false,"date":"2024-12-06","sourceType":"official","source":"cmmlu","sourceUrl":"https://github.com/haonan-li/CMMLU#排行榜","collected":"2026-10-10"}]}