[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"api:\u002Fnav:{}":3,"api:\u002Frecords\u002F16:{}":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",{"record":105,"variant":126,"benchmark":143,"rank":54,"ranked_count":72,"samples":151,"other_harnesses":155,"distribution":156,"cases":167},{"id":106,"notes":107,"score":108,"metrics":109,"cost_usd":107,"settings":110,"num_cases":118,"score_std":107,"harness_id":107,"has_bundle":119,"is_primary":119,"output_tps":107,"variant_id":120,"num_samples":54,"benchmark_id":54,"evaluated_at":121,"harness_name":107,"harness_slug":107,"input_tokens":122,"ttft_seconds":107,"output_tokens":123,"parent_run_id":107,"harness_version":107,"avg_case_seconds":124,"reasoning_tokens":107,"cost_per_case_usd":107,"trajectory_bundle_size_bytes":125},16,null,"0.5710",{},{"top_p":111,"recipe":107,"harness":107,"judge_model":112,"num_samples":54,"temperature":113,"extra_settings":114,"scoring_method":107,"harness_version":107,"framework_version":116,"max_output_tokens":117},"1.000","Qwen3-30B-A3B-Instruct-2507","0.60",{"top_k":115},-1,"OPT 1478395",65536,1000,true,3,"2026-10-26",118664,764396,"45.53",890172,{"access":127,"effort":128,"family":129,"is_tier":119,"hardware":107,"model_id":72,"provider":107,"data_date":5,"precision":107,"categories":130,"is_primary":119,"model_name":20,"model_slug":21,"short_name":20,"variant_id":120,"effort_rank":138,"input_price":107,"vendor_name":16,"vendor_slug":17,"architecture":107,"output_price":107,"overall_rank":54,"ranked_count":72,"release_date":139,"variant_name":140,"variant_slug":141,"vendor_order":142,"weights_repo":107,"missing_count":120,"overall_score":22,"context_window":107,"total_cost_usd":107,"total_params_b":107,"weights_source":107,"active_params_b":107,"cache_hit_price":107,"cost_incomplete":119,"is_open_weights":119,"inference_engine":107},"self_hosted","max","DeepSeek-V4",[131,133,135,137],{"name":41,"rank":72,"slug":42,"score":132,"expected":43,"available":54,"ranked_count":72},"2.857142857142860000000000",{"name":45,"rank":54,"slug":46,"score":134,"expected":47,"available":72,"ranked_count":72},"69.724166666666667500000000",{"name":49,"rank":120,"slug":50,"score":136,"expected":43,"available":54,"ranked_count":72},"45.670000000000000000000000",{"name":52,"rank":43,"slug":53,"score":136,"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,{"id":54,"name":99,"slug":100,"license":144,"metrics":145,"version":107,"is_active":119,"num_cases":118,"parent_id":107,"score_max":146,"score_min":147,"created_at":148,"is_agentic":4,"sort_order":142,"updated_at":148,"description":107,"metric_name":149,"homepage_url":107,"license_note":107,"scoring_method":107,"composite_weight":146,"higher_is_better":119,"bundle_size_bytes":150},"public",[],"1.0000","0.0000","2026-10-28T09:47:38.906402+08:00","Accuracy",110385750,[152],{"total":118,"failed":153,"passed":154,"run_id":106,"ungraded":142,"sample_idx":142,"trajectories":118},429,571,[],[157,158,161,164],{"name":20,"rank":54,"score":108,"run_id":106,"model_slug":21,"vendor_slug":17},{"name":28,"rank":43,"score":159,"run_id":160,"model_slug":29,"vendor_slug":25},"0.4460",61,{"name":36,"rank":120,"score":162,"run_id":163,"model_slug":37,"vendor_slug":33},"0.4330",43,{"name":12,"rank":72,"score":165,"run_id":166,"model_slug":13,"vendor_slug":9},"0.1110",119,{"restricted":4,"rows":168,"counts":489,"next_after":483},[169,178,184,191,198,205,212,219,226,233,239,246,253,259,266,272,279,285,291,298,304,311,318,325,331,337,343,349,355,362,368,374,381,386,392,398,404,410,416,422,428,434,440,446,452,459,464,470,476,482],{"id":170,"error":107,"extra":171,"score":146,"passed":119,"case_key":174,"cost_usd":107,"sample_idx":142,"input_tokens":175,"output_tokens":176,"has_trajectory":119,"duration_seconds":177},45055,{"status":172,"finish_reason":173},"completed","stop","0",121,347,"19.39",{"id":179,"error":107,"extra":180,"score":146,"passed":119,"case_key":181,"cost_usd":107,"sample_idx":142,"input_tokens":166,"output_tokens":182,"has_trajectory":119,"duration_seconds":183},45056,{"status":172,"finish_reason":173},"1",176,"14.50",{"id":185,"error":107,"extra":186,"score":146,"passed":119,"case_key":187,"cost_usd":107,"sample_idx":142,"input_tokens":188,"output_tokens":189,"has_trajectory":119,"duration_seconds":190},45057,{"status":172,"finish_reason":173},"2",118,217,"12.92",{"id":192,"error":107,"extra":193,"score":146,"passed":119,"case_key":194,"cost_usd":107,"sample_idx":142,"input_tokens":195,"output_tokens":196,"has_trajectory":119,"duration_seconds":197},45058,{"status":172,"finish_reason":173},"3",124,183,"14.78",{"id":199,"error":107,"extra":200,"score":146,"passed":119,"case_key":201,"cost_usd":107,"sample_idx":142,"input_tokens":202,"output_tokens":203,"has_trajectory":119,"duration_seconds":204},45059,{"status":172,"finish_reason":173},"4",133,277,"20.81",{"id":206,"error":107,"extra":207,"score":146,"passed":119,"case_key":208,"cost_usd":107,"sample_idx":142,"input_tokens":209,"output_tokens":210,"has_trajectory":119,"duration_seconds":211},45060,{"status":172,"finish_reason":173},"5",115,296,"20.43",{"id":213,"error":107,"extra":214,"score":147,"passed":4,"case_key":215,"cost_usd":107,"sample_idx":142,"input_tokens":216,"output_tokens":217,"has_trajectory":119,"duration_seconds":218},45061,{"status":172,"finish_reason":173},"6",116,368,"26.70",{"id":220,"error":107,"extra":221,"score":146,"passed":119,"case_key":222,"cost_usd":107,"sample_idx":142,"input_tokens":223,"output_tokens":224,"has_trajectory":119,"duration_seconds":225},45062,{"status":172,"finish_reason":173},"7",122,1863,"116.72",{"id":227,"error":107,"extra":228,"score":147,"passed":4,"case_key":229,"cost_usd":107,"sample_idx":142,"input_tokens":230,"output_tokens":231,"has_trajectory":119,"duration_seconds":232},45063,{"status":172,"finish_reason":173},"8",127,333,"14.64",{"id":234,"error":107,"extra":235,"score":147,"passed":4,"case_key":236,"cost_usd":107,"sample_idx":142,"input_tokens":175,"output_tokens":237,"has_trajectory":119,"duration_seconds":238},45064,{"status":172,"finish_reason":173},"9",432,"31.26",{"id":240,"error":107,"extra":241,"score":147,"passed":4,"case_key":242,"cost_usd":107,"sample_idx":142,"input_tokens":243,"output_tokens":244,"has_trajectory":119,"duration_seconds":245},45065,{"status":172,"finish_reason":173},"10",156,1631,"88.64",{"id":247,"error":107,"extra":248,"score":147,"passed":4,"case_key":249,"cost_usd":107,"sample_idx":142,"input_tokens":250,"output_tokens":251,"has_trajectory":119,"duration_seconds":252},45066,{"status":172,"finish_reason":173},"11",120,509,"29.81",{"id":254,"error":107,"extra":255,"score":147,"passed":4,"case_key":256,"cost_usd":107,"sample_idx":142,"input_tokens":216,"output_tokens":257,"has_trajectory":119,"duration_seconds":258},45067,{"status":172,"finish_reason":173},"12",135,"12.02",{"id":260,"error":107,"extra":261,"score":146,"passed":119,"case_key":262,"cost_usd":107,"sample_idx":142,"input_tokens":263,"output_tokens":264,"has_trajectory":119,"duration_seconds":265},45068,{"status":172,"finish_reason":173},"13",112,111,"7.38",{"id":267,"error":107,"extra":268,"score":146,"passed":119,"case_key":269,"cost_usd":107,"sample_idx":142,"input_tokens":270,"output_tokens":257,"has_trajectory":119,"duration_seconds":271},45069,{"status":172,"finish_reason":173},"14",110,"6.94",{"id":273,"error":107,"extra":274,"score":146,"passed":119,"case_key":275,"cost_usd":107,"sample_idx":142,"input_tokens":276,"output_tokens":277,"has_trajectory":119,"duration_seconds":278},45070,{"status":172,"finish_reason":173},"15",108,153,"8.96",{"id":280,"error":107,"extra":281,"score":146,"passed":119,"case_key":282,"cost_usd":107,"sample_idx":142,"input_tokens":264,"output_tokens":283,"has_trajectory":119,"duration_seconds":284},45071,{"status":172,"finish_reason":173},"16",146,"9.63",{"id":286,"error":107,"extra":287,"score":146,"passed":119,"case_key":288,"cost_usd":107,"sample_idx":142,"input_tokens":263,"output_tokens":289,"has_trajectory":119,"duration_seconds":290},45072,{"status":172,"finish_reason":173},"17",152,"9.86",{"id":292,"error":107,"extra":293,"score":147,"passed":4,"case_key":294,"cost_usd":107,"sample_idx":142,"input_tokens":295,"output_tokens":296,"has_trajectory":119,"duration_seconds":297},45073,{"status":172,"finish_reason":173},"18",142,1300,"85.45",{"id":299,"error":107,"extra":300,"score":146,"passed":119,"case_key":301,"cost_usd":107,"sample_idx":142,"input_tokens":250,"output_tokens":302,"has_trajectory":119,"duration_seconds":303},45074,{"status":172,"finish_reason":173},"19",144,"9.30",{"id":305,"error":107,"extra":306,"score":146,"passed":119,"case_key":307,"cost_usd":107,"sample_idx":142,"input_tokens":308,"output_tokens":309,"has_trajectory":119,"duration_seconds":310},45075,{"status":172,"finish_reason":173},"20",107,868,"61.62",{"id":312,"error":107,"extra":313,"score":147,"passed":4,"case_key":314,"cost_usd":107,"sample_idx":142,"input_tokens":315,"output_tokens":316,"has_trajectory":119,"duration_seconds":317},45076,{"status":172,"finish_reason":173},"21",117,1022,"75.51",{"id":319,"error":107,"extra":320,"score":146,"passed":119,"case_key":321,"cost_usd":107,"sample_idx":142,"input_tokens":322,"output_tokens":323,"has_trajectory":119,"duration_seconds":324},45077,{"status":172,"finish_reason":173},"22",114,180,"11.46",{"id":326,"error":107,"extra":327,"score":146,"passed":119,"case_key":328,"cost_usd":107,"sample_idx":142,"input_tokens":209,"output_tokens":329,"has_trajectory":119,"duration_seconds":330},45078,{"status":172,"finish_reason":173},"23",1717,"109.37",{"id":332,"error":107,"extra":333,"score":146,"passed":119,"case_key":334,"cost_usd":107,"sample_idx":142,"input_tokens":322,"output_tokens":335,"has_trajectory":119,"duration_seconds":336},45079,{"status":172,"finish_reason":173},"24",434,"24.60",{"id":338,"error":107,"extra":339,"score":146,"passed":119,"case_key":340,"cost_usd":107,"sample_idx":142,"input_tokens":270,"output_tokens":341,"has_trajectory":119,"duration_seconds":342},45080,{"status":172,"finish_reason":173},"25",281,"12.91",{"id":344,"error":107,"extra":345,"score":147,"passed":4,"case_key":346,"cost_usd":107,"sample_idx":142,"input_tokens":250,"output_tokens":347,"has_trajectory":119,"duration_seconds":348},45081,{"status":172,"finish_reason":173},"26",213,"11.05",{"id":350,"error":107,"extra":351,"score":146,"passed":119,"case_key":352,"cost_usd":107,"sample_idx":142,"input_tokens":223,"output_tokens":353,"has_trajectory":119,"duration_seconds":354},45082,{"status":172,"finish_reason":173},"27",1200,"72.26",{"id":356,"error":107,"extra":357,"score":146,"passed":119,"case_key":358,"cost_usd":107,"sample_idx":142,"input_tokens":359,"output_tokens":360,"has_trajectory":119,"duration_seconds":361},45083,{"status":172,"finish_reason":173},"28",123,491,"39.85",{"id":363,"error":107,"extra":364,"score":147,"passed":4,"case_key":365,"cost_usd":107,"sample_idx":142,"input_tokens":216,"output_tokens":366,"has_trajectory":119,"duration_seconds":367},45084,{"status":172,"finish_reason":173},"29",246,"16.11",{"id":369,"error":107,"extra":370,"score":147,"passed":4,"case_key":371,"cost_usd":107,"sample_idx":142,"input_tokens":359,"output_tokens":372,"has_trajectory":119,"duration_seconds":373},45085,{"status":172,"finish_reason":173},"30",939,"62.15",{"id":375,"error":107,"extra":376,"score":146,"passed":119,"case_key":377,"cost_usd":107,"sample_idx":142,"input_tokens":378,"output_tokens":379,"has_trajectory":119,"duration_seconds":380},45086,{"status":172,"finish_reason":173},"31",109,839,"43.85",{"id":382,"error":107,"extra":383,"score":146,"passed":119,"case_key":384,"cost_usd":107,"sample_idx":142,"input_tokens":315,"output_tokens":385,"has_trajectory":119,"duration_seconds":232},45087,{"status":172,"finish_reason":173},"32",199,{"id":387,"error":107,"extra":388,"score":147,"passed":4,"case_key":389,"cost_usd":107,"sample_idx":142,"input_tokens":378,"output_tokens":390,"has_trajectory":119,"duration_seconds":391},45088,{"status":172,"finish_reason":173},"33",394,"24.76",{"id":393,"error":107,"extra":394,"score":146,"passed":119,"case_key":395,"cost_usd":107,"sample_idx":142,"input_tokens":216,"output_tokens":396,"has_trajectory":119,"duration_seconds":397},45089,{"status":172,"finish_reason":173},"34",138,"10.45",{"id":399,"error":107,"extra":400,"score":146,"passed":119,"case_key":401,"cost_usd":107,"sample_idx":142,"input_tokens":359,"output_tokens":402,"has_trajectory":119,"duration_seconds":403},45090,{"status":172,"finish_reason":173},"35",350,"21.03",{"id":405,"error":107,"extra":406,"score":147,"passed":4,"case_key":407,"cost_usd":107,"sample_idx":142,"input_tokens":166,"output_tokens":408,"has_trajectory":119,"duration_seconds":409},45091,{"status":172,"finish_reason":173},"36",527,"39.60",{"id":411,"error":107,"extra":412,"score":146,"passed":119,"case_key":413,"cost_usd":107,"sample_idx":142,"input_tokens":315,"output_tokens":414,"has_trajectory":119,"duration_seconds":415},45092,{"status":172,"finish_reason":173},"37",220,"15.28",{"id":417,"error":107,"extra":418,"score":146,"passed":119,"case_key":419,"cost_usd":107,"sample_idx":142,"input_tokens":264,"output_tokens":420,"has_trajectory":119,"duration_seconds":421},45093,{"status":172,"finish_reason":173},"38",263,"16.02",{"id":423,"error":107,"extra":424,"score":146,"passed":119,"case_key":425,"cost_usd":107,"sample_idx":142,"input_tokens":263,"output_tokens":426,"has_trajectory":119,"duration_seconds":427},45094,{"status":172,"finish_reason":173},"39",445,"33.95",{"id":429,"error":107,"extra":430,"score":146,"passed":119,"case_key":431,"cost_usd":107,"sample_idx":142,"input_tokens":264,"output_tokens":432,"has_trajectory":119,"duration_seconds":433},45095,{"status":172,"finish_reason":173},"40",159,"10.83",{"id":435,"error":107,"extra":436,"score":146,"passed":119,"case_key":437,"cost_usd":107,"sample_idx":142,"input_tokens":315,"output_tokens":438,"has_trajectory":119,"duration_seconds":439},45096,{"status":172,"finish_reason":173},"41",134,"8.86",{"id":441,"error":107,"extra":442,"score":146,"passed":119,"case_key":443,"cost_usd":107,"sample_idx":142,"input_tokens":322,"output_tokens":444,"has_trajectory":119,"duration_seconds":445},45097,{"status":172,"finish_reason":173},"42",207,"16.41",{"id":447,"error":107,"extra":448,"score":147,"passed":4,"case_key":449,"cost_usd":107,"sample_idx":142,"input_tokens":166,"output_tokens":450,"has_trajectory":119,"duration_seconds":451},45098,{"status":172,"finish_reason":173},"43",832,"52.67",{"id":453,"error":107,"extra":454,"score":146,"passed":119,"case_key":455,"cost_usd":107,"sample_idx":142,"input_tokens":456,"output_tokens":457,"has_trajectory":119,"duration_seconds":458},45099,{"status":172,"finish_reason":173},"44",126,148,"9.61",{"id":460,"error":107,"extra":461,"score":147,"passed":4,"case_key":462,"cost_usd":107,"sample_idx":142,"input_tokens":264,"output_tokens":251,"has_trajectory":119,"duration_seconds":463},45100,{"status":172,"finish_reason":173},"45","30.08",{"id":465,"error":107,"extra":466,"score":147,"passed":4,"case_key":467,"cost_usd":107,"sample_idx":142,"input_tokens":315,"output_tokens":468,"has_trajectory":119,"duration_seconds":469},45101,{"status":172,"finish_reason":173},"46",226,"14.32",{"id":471,"error":107,"extra":472,"score":146,"passed":119,"case_key":473,"cost_usd":107,"sample_idx":142,"input_tokens":308,"output_tokens":474,"has_trajectory":119,"duration_seconds":475},45102,{"status":172,"finish_reason":173},"47",576,"28.64",{"id":477,"error":107,"extra":478,"score":147,"passed":4,"case_key":479,"cost_usd":107,"sample_idx":142,"input_tokens":276,"output_tokens":480,"has_trajectory":119,"duration_seconds":481},45103,{"status":172,"finish_reason":173},"48",141,"10.19",{"id":483,"error":107,"extra":484,"score":147,"passed":4,"case_key":485,"cost_usd":107,"sample_idx":142,"input_tokens":486,"output_tokens":487,"has_trajectory":119,"duration_seconds":488},45104,{"status":172,"finish_reason":173},"49",137,480,"23.11",{"total":118,"failed":153,"passed":154,"run_id":106,"ungraded":142,"sample_idx":142,"trajectories":118}]