[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"api:\u002Fnav:{}":3,"api:\u002Fleaderboard:{\"effort\":\"top\",\"weights\":\"all\",\"sort\":\"overall\",\"limit\":100,\"page\":1}":104},{"login_code_required":4,"data_date":5,"vendors":6,"categories":39,"topics":55,"model_count":72,"benchmark_count":73,"models":74,"benchmarks":79},false,"2026-10-27",[7,15,23,31],{"name":8,"slug":9,"models":10},"阿里巴巴","alibaba",[11],{"name":12,"slug":13,"short_name":12,"overall_score":14},"Qwen3.8-27B","qwen3-8-27b","47.576875000000000000000000",{"name":16,"slug":17,"models":18},"深度求索","deepseek",[19],{"name":20,"slug":21,"short_name":20,"overall_score":22},"DeepSeek-V4-Pro","deepseek-v4-pro","65.030000000000000000000000",{"name":24,"slug":25,"models":26},"月之暗面","moonshot",[27],{"name":28,"slug":29,"short_name":28,"overall_score":30},"Kimi-K3","kimi-k3","57.474000000000000000000000",{"name":32,"slug":33,"models":34},"智谱","zhipu",[35],{"name":36,"slug":37,"short_name":36,"overall_score":38},"GLM-5.3","glm-5-3","61.067500000000000000000000",[40,44,48,51],{"name":41,"slug":42,"benchmark_count":43},"体验感","experience",2,{"name":45,"slug":46,"benchmark_count":47},"认知能力","cognition",5,{"name":49,"slug":50,"benchmark_count":43},"Agentic 能力","agentic",{"name":52,"slug":53,"benchmark_count":54},"应用领域","domain",1,[56,60,64,68],{"slug":57,"name":58,"question":59},"quant","量化 vs 原精度","量化之后能力掉多少",{"slug":61,"name":62,"question":63},"selfhost","自部署 vs 官方 API","同一份权重，自己部署和调官方 API，分数一致吗",{"slug":65,"name":66,"question":67},"size","同家族尺寸","同一代模型，小尺寸比旗舰差多少",{"slug":69,"name":70,"question":71},"harness","多 Harness","同一个模型换一个 harness，agentic 分数变多少",4,8,[75,76,77,78],{"name":20,"slug":21},{"name":36,"slug":37},{"name":28,"slug":29},{"name":12,"slug":13},[80,83,86,89,92,95,98,101],{"name":81,"slug":82},"AA-LCR","aa-lcr",{"name":84,"slug":85},"AA-Omniscience","aa-omniscience",{"name":87,"slug":88},"AutomationBench","automationbench",{"name":90,"slug":91},"CL-bench","cl-bench",{"name":93,"slug":94},"HLE","hle",{"name":96,"slug":97},"MMLU-Pro","mmlu-pro",{"name":99,"slug":100},"SimpleQA Verified","simpleqa-verified",{"name":102,"slug":103},"SWE-bench Pro","swe-bench-pro",{"total":72,"rows":105,"categories":305,"columns":310,"ranges":329},[106,164,210,252],{"access":107,"effort":108,"family":109,"is_tier":110,"hardware":111,"model_id":72,"provider":111,"data_date":5,"precision":111,"benchmarks":112,"categories":151,"is_primary":110,"model_name":20,"model_slug":21,"short_name":20,"variant_id":122,"effort_rank":159,"input_price":111,"vendor_name":16,"vendor_slug":17,"architecture":111,"output_price":111,"overall_rank":54,"ranked_count":72,"release_date":160,"variant_name":161,"variant_slug":162,"vendor_order":163,"weights_repo":111,"missing_count":122,"overall_score":22,"category_score":111,"context_window":111,"total_cost_usd":111,"total_params_b":111,"weights_source":111,"active_params_b":111,"benchmark_score":111,"cache_hit_price":111,"cost_incomplete":110,"is_open_weights":110,"inference_engine":111},"self_hosted","max","DeepSeek-V4",true,null,[113,118,120,126,130,132,137,142,147,149],{"key":114,"rank":111,"slug":85,"score":115,"run_id":116,"category_slug":42,"normalized_score":117},"experience:aa-omniscience","0.9714285714285714",13,"2.85714285714286000000",{"key":119,"rank":111,"slug":91,"score":111,"run_id":111,"category_slug":42,"normalized_score":111},"experience:cl-bench",{"key":121,"rank":122,"slug":82,"score":123,"run_id":124,"category_slug":46,"normalized_score":125},"cognition:aa-lcr",3,"0.7550",12,"75.50000000000000000000",{"key":127,"rank":111,"slug":85,"score":128,"run_id":116,"category_slug":46,"normalized_score":129},"cognition:aa-omniscience","0.5916666666666667","59.16666666666667000000",{"key":131,"rank":111,"slug":94,"score":111,"run_id":111,"category_slug":46,"normalized_score":111},"cognition:hle",{"key":133,"rank":43,"slug":97,"score":134,"run_id":135,"category_slug":46,"normalized_score":136},"cognition:mmlu-pro","0.8713",15,"87.13000000000000000000",{"key":138,"rank":54,"slug":100,"score":139,"run_id":140,"category_slug":46,"normalized_score":141},"cognition:simpleqa-verified","0.5710",16,"57.10000000000000000000",{"key":143,"rank":43,"slug":88,"score":144,"run_id":145,"category_slug":50,"normalized_score":146},"agentic:automationbench","0.4567",14,"45.67000000000000000000",{"key":148,"rank":111,"slug":103,"score":111,"run_id":111,"category_slug":50,"normalized_score":111},"agentic:swe-bench-pro",{"key":150,"rank":43,"slug":88,"score":144,"run_id":145,"category_slug":53,"normalized_score":146},"domain:automationbench",[152,154,156,158],{"name":41,"rank":72,"slug":42,"score":153,"expected":43,"available":54,"ranked_count":72},"2.857142857142860000000000",{"name":45,"rank":54,"slug":46,"score":155,"expected":47,"available":72,"ranked_count":72},"69.724166666666667500000000",{"name":49,"rank":122,"slug":50,"score":157,"expected":43,"available":54,"ranked_count":72},"45.670000000000000000000000",{"name":52,"rank":43,"slug":53,"score":157,"expected":54,"available":54,"ranked_count":72},6,"2026-08-13","DeepSeek-V4-Pro 自部署 max","deepseek-v4-pro-self-hosted-max",0,{"access":165,"effort":108,"family":36,"is_tier":110,"hardware":111,"model_id":54,"provider":111,"data_date":166,"precision":111,"benchmarks":167,"categories":198,"is_primary":110,"model_name":36,"model_slug":37,"short_name":36,"variant_id":73,"effort_rank":159,"input_price":111,"vendor_name":32,"vendor_slug":33,"architecture":111,"output_price":111,"overall_rank":43,"ranked_count":72,"release_date":207,"variant_name":208,"variant_slug":209,"vendor_order":163,"weights_repo":111,"missing_count":43,"overall_score":38,"category_score":111,"context_window":111,"total_cost_usd":111,"total_params_b":111,"weights_source":111,"active_params_b":111,"benchmark_score":111,"cache_hit_price":111,"cost_incomplete":110,"is_open_weights":110,"inference_engine":111},"api","2026-10-23",[168,172,173,177,180,181,185,189,193,197],{"key":114,"rank":111,"slug":85,"score":169,"run_id":170,"category_slug":42,"normalized_score":171},"0.27002967359050445",40,"72.99703264094955500000",{"key":119,"rank":111,"slug":91,"score":111,"run_id":111,"category_slug":42,"normalized_score":111},{"key":121,"rank":72,"slug":82,"score":174,"run_id":175,"category_slug":46,"normalized_score":176},"0.6525",39,"65.25000000000000000000",{"key":127,"rank":111,"slug":85,"score":178,"run_id":170,"category_slug":46,"normalized_score":179},"0.43833333333333335","43.83333333333333500000",{"key":131,"rank":111,"slug":94,"score":111,"run_id":111,"category_slug":46,"normalized_score":111},{"key":133,"rank":72,"slug":97,"score":182,"run_id":183,"category_slug":46,"normalized_score":184},"0.8507",42,"85.07000000000000000000",{"key":138,"rank":122,"slug":100,"score":186,"run_id":187,"category_slug":46,"normalized_score":188},"0.4330",43,"43.30000000000000000000",{"key":143,"rank":54,"slug":88,"score":190,"run_id":191,"category_slug":50,"normalized_score":192},"0.5250",41,"52.50000000000000000000",{"key":148,"rank":43,"slug":103,"score":194,"run_id":195,"category_slug":50,"normalized_score":196},"0.5595",44,"55.95000000000000000000",{"key":150,"rank":54,"slug":88,"score":190,"run_id":191,"category_slug":53,"normalized_score":192},[199,201,203,205],{"name":41,"rank":54,"slug":42,"score":200,"expected":43,"available":54,"ranked_count":72},"72.997032640949555000000000",{"name":45,"rank":122,"slug":46,"score":202,"expected":47,"available":72,"ranked_count":72},"59.363333333333333750000000",{"name":49,"rank":54,"slug":50,"score":204,"expected":43,"available":43,"ranked_count":72},"54.225000000000000000000000",{"name":52,"rank":54,"slug":53,"score":206,"expected":54,"available":54,"ranked_count":72},"52.500000000000000000000000","2026-08-18","GLM-5.3 官方 API max","glm-5-3-api-max",{"access":165,"effort":108,"family":28,"is_tier":110,"hardware":111,"model_id":122,"provider":111,"data_date":211,"precision":111,"benchmarks":212,"categories":240,"is_primary":110,"model_name":28,"model_slug":29,"short_name":28,"variant_id":248,"effort_rank":159,"input_price":111,"vendor_name":24,"vendor_slug":25,"architecture":111,"output_price":111,"overall_rank":122,"ranked_count":72,"release_date":249,"variant_name":250,"variant_slug":251,"vendor_order":163,"weights_repo":111,"missing_count":122,"overall_score":30,"category_score":111,"context_window":111,"total_cost_usd":111,"total_params_b":111,"weights_source":111,"active_params_b":111,"benchmark_score":111,"cache_hit_price":111,"cost_incomplete":110,"is_open_weights":110,"inference_engine":111},"2026-10-24",[213,217,218,222,225,226,230,234,238,239],{"key":114,"rank":111,"slug":85,"score":214,"run_id":215,"category_slug":42,"normalized_score":216},"0.3494809688581315",58,"65.05190311418685000000",{"key":119,"rank":111,"slug":91,"score":111,"run_id":111,"category_slug":42,"normalized_score":111},{"key":121,"rank":54,"slug":82,"score":219,"run_id":220,"category_slug":46,"normalized_score":221},"0.8300",57,"83.00000000000000000000",{"key":127,"rank":111,"slug":85,"score":223,"run_id":215,"category_slug":46,"normalized_score":224},"0.5183333333333333","51.83333333333333000000",{"key":131,"rank":111,"slug":94,"score":111,"run_id":111,"category_slug":46,"normalized_score":111},{"key":133,"rank":54,"slug":97,"score":227,"run_id":228,"category_slug":46,"normalized_score":229},"0.8877",60,"88.77000000000000000000",{"key":138,"rank":43,"slug":100,"score":231,"run_id":232,"category_slug":46,"normalized_score":233},"0.4460",61,"44.60000000000000000000",{"key":143,"rank":72,"slug":88,"score":235,"run_id":236,"category_slug":50,"normalized_score":237},"0.0350",59,"3.50000000000000000000",{"key":148,"rank":111,"slug":103,"score":111,"run_id":111,"category_slug":50,"normalized_score":111},{"key":150,"rank":72,"slug":88,"score":235,"run_id":236,"category_slug":53,"normalized_score":237},[241,243,245,247],{"name":41,"rank":43,"slug":42,"score":242,"expected":43,"available":54,"ranked_count":72},"65.051903114186850000000000",{"name":45,"rank":43,"slug":46,"score":244,"expected":47,"available":72,"ranked_count":72},"67.050833333333332500000000",{"name":49,"rank":72,"slug":50,"score":246,"expected":43,"available":54,"ranked_count":72},"3.500000000000000000000000",{"name":52,"rank":72,"slug":53,"score":246,"expected":54,"available":54,"ranked_count":72},11,"2026-07-16","Kimi-K3 官方 API max","kimi-k3-api-max",{"access":107,"effort":253,"family":254,"is_tier":110,"hardware":111,"model_id":43,"provider":111,"data_date":255,"precision":111,"benchmarks":256,"categories":293,"is_primary":110,"model_name":12,"model_slug":13,"short_name":12,"variant_id":302,"effort_rank":47,"input_price":111,"vendor_name":8,"vendor_slug":9,"architecture":111,"output_price":111,"overall_rank":72,"ranked_count":72,"release_date":111,"variant_name":303,"variant_slug":304,"vendor_order":163,"weights_repo":111,"missing_count":163,"overall_score":14,"category_score":111,"context_window":111,"total_cost_usd":111,"total_params_b":111,"weights_source":111,"active_params_b":111,"benchmark_score":111,"cache_hit_price":111,"cost_incomplete":110,"is_open_weights":110,"inference_engine":111},"xhigh","Qwen3.8","2026-10-22",[257,261,265,269,272,276,280,284,288,292],{"key":114,"rank":111,"slug":85,"score":258,"run_id":259,"category_slug":42,"normalized_score":260},"0.27983539094650206",113,"72.01646090534979400000",{"key":119,"rank":54,"slug":91,"score":262,"run_id":263,"category_slug":42,"normalized_score":264},"0.2489",115,"24.89000000000000000000",{"key":121,"rank":43,"slug":82,"score":266,"run_id":267,"category_slug":46,"normalized_score":268},"0.8125",112,"81.25000000000000000000",{"key":127,"rank":111,"slug":85,"score":270,"run_id":259,"category_slug":46,"normalized_score":271},"0.19","19.00000000000000000000",{"key":131,"rank":54,"slug":94,"score":273,"run_id":274,"category_slug":46,"normalized_score":275},"0.3109",116,"31.09000000000000000000",{"key":133,"rank":122,"slug":97,"score":277,"run_id":278,"category_slug":46,"normalized_score":279},"0.8516",117,"85.16000000000000000000",{"key":138,"rank":72,"slug":100,"score":281,"run_id":282,"category_slug":46,"normalized_score":283},"0.1110",119,"11.10000000000000000000",{"key":143,"rank":122,"slug":88,"score":285,"run_id":286,"category_slug":50,"normalized_score":287},"0.4150",114,"41.50000000000000000000",{"key":148,"rank":54,"slug":103,"score":289,"run_id":290,"category_slug":50,"normalized_score":291},"0.5746",120,"57.46000000000000000000",{"key":150,"rank":122,"slug":88,"score":285,"run_id":286,"category_slug":53,"normalized_score":287},[294,296,298,300],{"name":41,"rank":122,"slug":42,"score":295,"expected":43,"available":43,"ranked_count":72},"48.453230452674897000000000",{"name":45,"rank":72,"slug":46,"score":297,"expected":47,"available":47,"ranked_count":72},"45.520000000000000000000000",{"name":49,"rank":43,"slug":50,"score":299,"expected":43,"available":43,"ranked_count":72},"49.480000000000000000000000",{"name":52,"rank":122,"slug":53,"score":301,"expected":54,"available":54,"ranked_count":72},"41.500000000000000000000000",19,"Qwen3.8-27B 自部署 xhigh","qwen3-8-27b-self-hosted-xhigh",[306,307,308,309],{"name":41,"slug":42},{"name":45,"slug":46},{"name":49,"slug":50},{"name":52,"slug":53},[311,316,318,319,321,322,323,324,326,328],{"key":114,"name":84,"slug":85,"metric":312,"score_max":313,"score_min":314,"metric_name":315,"category_name":41,"category_slug":42,"higher_is_better":4},"hallucination_rate","1.0000","0.0000","Hallucination Rate",{"key":119,"name":90,"slug":91,"metric":111,"score_max":313,"score_min":314,"metric_name":317,"category_name":41,"category_slug":42,"higher_is_better":110},"Accuracy",{"key":121,"name":81,"slug":82,"metric":111,"score_max":313,"score_min":314,"metric_name":317,"category_name":45,"category_slug":46,"higher_is_better":110},{"key":127,"name":84,"slug":85,"metric":320,"score_max":313,"score_min":314,"metric_name":317,"category_name":45,"category_slug":46,"higher_is_better":110},"accuracy",{"key":131,"name":93,"slug":94,"metric":111,"score_max":313,"score_min":314,"metric_name":317,"category_name":45,"category_slug":46,"higher_is_better":110},{"key":133,"name":96,"slug":97,"metric":111,"score_max":313,"score_min":314,"metric_name":317,"category_name":45,"category_slug":46,"higher_is_better":110},{"key":138,"name":99,"slug":100,"metric":111,"score_max":313,"score_min":314,"metric_name":317,"category_name":45,"category_slug":46,"higher_is_better":110},{"key":143,"name":87,"slug":88,"metric":111,"score_max":313,"score_min":314,"metric_name":325,"category_name":49,"category_slug":50,"higher_is_better":110},"Pass Rate",{"key":148,"name":102,"slug":103,"metric":111,"score_max":313,"score_min":314,"metric_name":327,"category_name":49,"category_slug":50,"higher_is_better":110},"Resolved Rate",{"key":150,"name":87,"slug":88,"metric":111,"score_max":313,"score_min":314,"metric_name":325,"category_name":52,"category_slug":53,"higher_is_better":110},{"overall":330,"benchmarks":331,"categories":342},{"max":22,"min":14},{"cognition:hle":332,"cognition:aa-lcr":333,"cognition:mmlu-pro":334,"experience:cl-bench":335,"agentic:swe-bench-pro":336,"domain:automationbench":337,"agentic:automationbench":338,"cognition:aa-omniscience":339,"experience:aa-omniscience":340,"cognition:simpleqa-verified":341},{"max":275,"min":275},{"max":221,"min":176},{"max":229,"min":184},{"max":264,"min":264},{"max":291,"min":196},{"max":192,"min":237},{"max":192,"min":237},{"max":129,"min":271},{"max":171,"min":117},{"max":141,"min":283},{"domain":343,"agentic":344,"cognition":345,"experience":346},{"max":206,"min":246},{"max":204,"min":246},{"max":155,"min":297},{"max":200,"min":153}]