{"meta":{"generatedAt":"2026-11-01T19:55:06Z","site":"https://models.star365.site","notice":"점수는 각 원본의 측정치이며 star365 가 독자 측정한 값이 아니다. 빈 값은 미보고이며 추정하지 않는다.","sources":[{"id":"aa-v5","name":"Artificial Analysis Intelligence Index v5.0","url":"https://artificialanalysis.ai/leaderboards/models","fetchedAt":"2026-11-01T19:55:04Z"},{"id":"all-bench","name":"ALL-Bench v2.1","url":"https://github.com/final-bench/ALL-Bench-Leaderboard","lastVerified":"2026-03-08","status":"원본 2026-03-08 이후 갱신 중단"}],"weightedFormula":"weighted = mean(scores) × sqrt(min(N,10)/10)","license":"원본 데이터의 라이선스를 따른다. AA 지표는 출처 표기 필요."},"aa":{"benchmark":"Artificial Analysis Intelligence Index v5.0","fetchedAt":"2026-11-01T19:55:04Z","models":[{"slug":"gemini-4-carbon","name":"Gemini 4 Carbon","creator":"Google","context":1000000,"released":"2026-10-28","index":59.0,"bestVariant":"high","costPerTask":7.76,"tokPerSec":85.0,"ttft":118.58,"e2e":124.49,"variants":[{"variant":"high","index":59.0,"costPerTask":7.76,"tokPerSec":85.0,"ttft":118.58,"e2e":124.49}]},{"slug":"claude-opus-5-5","name":"Claude Opus 5.5","creator":"Anthropic","context":1000000,"released":"2026-09-22","index":58.0,"bestVariant":"max with fallback","costPerTask":10.04,"tokPerSec":97.0,"ttft":729.62,"e2e":734.79,"variants":[{"variant":"max with fallback","index":58.0,"costPerTask":10.04,"tokPerSec":97.0,"ttft":729.62,"e2e":734.79},{"variant":"xhigh with fallback","index":57.0,"costPerTask":6.05,"tokPerSec":85.0,"ttft":127.23,"e2e":133.08},{"variant":"high with fallback","index":52.0,"costPerTask":3.46,"tokPerSec":80.0,"ttft":34.52,"e2e":40.8},{"variant":"medium with fallback","index":48.0,"costPerTask":2.45,"tokPerSec":80.0,"ttft":16.41,"e2e":22.69},{"variant":"low with fallback","index":35.0,"costPerTask":0.98,"tokPerSec":78.0,"ttft":6.16,"e2e":12.6}]},{"slug":"gpt-6-astra","name":"GPT-6 Astra","creator":"OpenAI","context":1000000,"released":null,"index":57.0,"bestVariant":"max","costPerTask":6.24,"tokPerSec":78.0,"ttft":270.03,"e2e":276.42,"variants":[{"variant":"max","index":57.0,"costPerTask":6.24,"tokPerSec":78.0,"ttft":270.03,"e2e":276.42},{"variant":"xhigh","index":55.0,"costPerTask":3.66,"tokPerSec":66.0,"ttft":149.41,"e2e":156.99},{"variant":"high","index":55.0,"costPerTask":3.24,"tokPerSec":63.0,"ttft":48.12,"e2e":56.08},{"variant":"medium","index":51.0,"costPerTask":2.73,"tokPerSec":57.0,"ttft":5.07,"e2e":13.86},{"variant":"low","index":43.0,"costPerTask":1.42,"tokPerSec":58.0,"ttft":2.02,"e2e":10.71}]},{"slug":"gpt-6-1-sol","name":"GPT-6.1 Sol","creator":"OpenAI","context":1000000,"released":null,"index":55.0,"bestVariant":"max","costPerTask":1.27,"tokPerSec":79.0,"ttft":315.06,"e2e":321.37,"variants":[{"variant":"max","index":55.0,"costPerTask":1.27,"tokPerSec":79.0,"ttft":315.06,"e2e":321.37},{"variant":"xhigh","index":52.0,"costPerTask":0.76,"tokPerSec":71.0,"ttft":126.08,"e2e":133.12},{"variant":"high","index":50.0,"costPerTask":0.59,"tokPerSec":60.0,"ttft":44.36,"e2e":52.7},{"variant":"medium","index":46.0,"costPerTask":0.36,"tokPerSec":50.0,"ttft":4.99,"e2e":15.04},{"variant":"low","index":35.0,"costPerTask":0.16,"tokPerSec":52.0,"ttft":2.28,"e2e":11.84}]},{"slug":"claude-sonnet-5-5","name":"Claude Sonnet 5.5","creator":"Anthropic","context":1000000,"released":"2026-09-28","index":54.0,"bestVariant":"max with fallback","costPerTask":10.02,"tokPerSec":139.0,"ttft":463.03,"e2e":466.62,"variants":[{"variant":"max with fallback","index":54.0,"costPerTask":10.02,"tokPerSec":139.0,"ttft":463.03,"e2e":466.62},{"variant":"xhigh with fallback","index":50.0,"costPerTask":4.09,"tokPerSec":107.0,"ttft":29.68,"e2e":34.34},{"variant":"high with fallback","index":41.0,"costPerTask":1.64,"tokPerSec":103.0,"ttft":10.95,"e2e":15.8},{"variant":"medium with fallback","index":30.0,"costPerTask":0.78,"tokPerSec":101.0,"ttft":3.75,"e2e":8.71},{"variant":"low with fallback","index":25.0,"costPerTask":0.51,"tokPerSec":105.0,"ttft":1.3,"e2e":6.04}]},{"slug":"gpt-6-1-terra","name":"GPT-6.1 Terra","creator":"OpenAI","context":1000000,"released":"2026-10-15","index":51.0,"bestVariant":"max","costPerTask":0.96,"tokPerSec":97.0,"ttft":274.62,"e2e":279.76,"variants":[{"variant":"max","index":51.0,"costPerTask":0.96,"tokPerSec":97.0,"ttft":274.62,"e2e":279.76},{"variant":"xhigh","index":45.0,"costPerTask":0.4,"tokPerSec":92.0,"ttft":45.76,"e2e":51.19},{"variant":"high","index":42.0,"costPerTask":0.26,"tokPerSec":91.0,"ttft":12.98,"e2e":18.48},{"variant":"medium","index":38.0,"costPerTask":0.17,"tokPerSec":91.0,"ttft":3.11,"e2e":8.61},{"variant":"non-reasoning","index":30.0,"costPerTask":0.23,"tokPerSec":90.0,"ttft":0.96,"e2e":6.52},{"variant":"low","index":29.0,"costPerTask":0.07,"tokPerSec":91.0,"ttft":1.41,"e2e":6.91}]},{"slug":"claude-fable-5-1","name":"Claude Fable 5.1","creator":"Anthropic","context":1000000,"released":"2026-09-01","index":50.0,"bestVariant":"xhigh with fallback","costPerTask":10.85,"tokPerSec":62.0,"ttft":73.47,"e2e":81.51,"variants":[{"variant":"xhigh with fallback","index":50.0,"costPerTask":10.85,"tokPerSec":62.0,"ttft":73.47,"e2e":81.51},{"variant":"max with fallback","index":49.0,"costPerTask":12.99,"tokPerSec":69.0,"ttft":238.66,"e2e":245.94},{"variant":"high with fallback","index":47.0,"costPerTask":7.54,"tokPerSec":57.0,"ttft":14.62,"e2e":23.37},{"variant":"medium with fallback","index":43.0,"costPerTask":5.56,"tokPerSec":55.0,"ttft":3.25,"e2e":12.38},{"variant":"low with fallback","index":38.0,"costPerTask":4.47,"tokPerSec":55.0,"ttft":2.27,"e2e":11.39}]},{"slug":"gemini-4-argon","name":"Gemini 4 Argon","creator":"Google","context":1000000,"released":null,"index":50.0,"bestVariant":"high","costPerTask":7.68,"tokPerSec":83.0,"ttft":49.13,"e2e":55.15,"variants":[{"variant":"high","index":50.0,"costPerTask":7.68,"tokPerSec":83.0,"ttft":49.13,"e2e":55.15}]},{"slug":"muse-wonder-full","name":"Muse Wonder Full","creator":"Meta","context":1000000,"released":null,"index":50.0,"bestVariant":"max","costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"max","index":50.0,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"kimi-k3-1-pro","name":"Kimi K3.1 Pro","creator":"Kimi","context":1000000,"released":"2026-10-23","index":46.0,"bestVariant":null,"costPerTask":3.68,"tokPerSec":31.0,"ttft":3.27,"e2e":84.23,"variants":[{"variant":"","index":46.0,"costPerTask":3.68,"tokPerSec":31.0,"ttft":3.27,"e2e":84.23}]},{"slug":"muse-wonder","name":"Muse Wonder","creator":"Meta","context":1050000,"released":null,"index":43.0,"bestVariant":"max","costPerTask":6.95,"tokPerSec":143.0,"ttft":30.09,"e2e":47.56,"variants":[{"variant":"max","index":43.0,"costPerTask":6.95,"tokPerSec":143.0,"ttft":30.09,"e2e":47.56}]},{"slug":"claude-haiku-5-5","name":"Claude Haiku 5.5","creator":"Anthropic","context":1000000,"released":null,"index":36.0,"bestVariant":"max","costPerTask":0.41,"tokPerSec":235.0,"ttft":341.05,"e2e":343.18,"variants":[{"variant":"max","index":36.0,"costPerTask":0.41,"tokPerSec":235.0,"ttft":341.05,"e2e":343.18},{"variant":"xhigh","index":34.0,"costPerTask":0.25,"tokPerSec":186.0,"ttft":73.3,"e2e":75.98},{"variant":"high","index":30.0,"costPerTask":0.16,"tokPerSec":173.0,"ttft":19.76,"e2e":22.65},{"variant":"medium","index":25.0,"costPerTask":0.09,"tokPerSec":169.0,"ttft":11.41,"e2e":14.36},{"variant":"low","index":19.0,"costPerTask":0.04,"tokPerSec":158.0,"ttft":7.89,"e2e":11.05}]},{"slug":"grok-4-7","name":"Grok 4.7","creator":"SpaceXAI","context":500000,"released":null,"index":35.0,"bestVariant":"high","costPerTask":6.18,"tokPerSec":78.0,"ttft":32.59,"e2e":38.96,"variants":[{"variant":"high","index":35.0,"costPerTask":6.18,"tokPerSec":78.0,"ttft":32.59,"e2e":38.96},{"variant":"xhigh","index":34.0,"costPerTask":7.51,"tokPerSec":87.0,"ttft":42.06,"e2e":47.81},{"variant":"low","index":31.0,"costPerTask":2.92,"tokPerSec":72.0,"ttft":5.74,"e2e":12.68}]},{"slug":"qwen3-8-max","name":"Qwen3.8 Max","creator":"Alibaba","context":984000,"released":null,"index":34.0,"bestVariant":"0902","costPerTask":10.02,"tokPerSec":40.0,"ttft":3.1,"e2e":65.5,"variants":[{"variant":"0902","index":34.0,"costPerTask":10.02,"tokPerSec":40.0,"ttft":3.1,"e2e":65.5}]},{"slug":"deepseek-v4-1-flash-1017","name":"DeepSeek V4.1 Flash 1017","creator":"DeepSeek","context":1000000,"released":"2026-10-17","index":32.0,"bestVariant":"max","costPerTask":0.6,"tokPerSec":175.0,"ttft":1.1,"e2e":15.38,"variants":[{"variant":"max","index":32.0,"costPerTask":0.6,"tokPerSec":175.0,"ttft":1.1,"e2e":15.38},{"variant":"non-reasoning","index":19.0,"costPerTask":null,"tokPerSec":1486.0,"ttft":1.15,"e2e":1.49}]},{"slug":"glm-5-3","name":"GLM-5.3","creator":"Z AI","context":1000000,"released":null,"index":32.0,"bestVariant":"max","costPerTask":5.01,"tokPerSec":76.0,"ttft":1.76,"e2e":34.76,"variants":[{"variant":"max","index":32.0,"costPerTask":5.01,"tokPerSec":76.0,"ttft":1.76,"e2e":34.76},{"variant":"low","index":23.0,"costPerTask":1.57,"tokPerSec":72.0,"ttft":1.6,"e2e":36.28}]},{"slug":"gemini-3-8-flash","name":"Gemini 3.8 Flash","creator":"Google","context":1000000,"released":null,"index":31.0,"bestVariant":"high","costPerTask":7.77,"tokPerSec":120.0,"ttft":24.69,"e2e":28.85,"variants":[{"variant":"high","index":31.0,"costPerTask":7.77,"tokPerSec":120.0,"ttft":24.69,"e2e":28.85},{"variant":"medium","index":30.0,"costPerTask":6.26,"tokPerSec":null,"ttft":null,"e2e":null},{"variant":"low","index":22.0,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"mimo-v2-6-pro","name":"MiMo-V2.6-Pro","creator":"Xiaomi","context":1000000,"released":"2026-09-22","index":30.0,"bestVariant":null,"costPerTask":0.26,"tokPerSec":49.0,"ttft":3.09,"e2e":53.75,"variants":[{"variant":"","index":30.0,"costPerTask":0.26,"tokPerSec":49.0,"ttft":3.09,"e2e":53.75}]},{"slug":"ling-3-1-flash-1019","name":"Ling 3.1 Flash 1019","creator":"InclusionAI","context":1050000,"released":null,"index":29.0,"bestVariant":null,"costPerTask":3.19,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":29.0,"costPerTask":3.19,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"minimax-m3-1-flash","name":"MiniMax-M3.1-Flash","creator":"MiniMax","context":1000000,"released":"2026-10-19","index":29.0,"bestVariant":null,"costPerTask":0.49,"tokPerSec":119.0,"ttft":1.28,"e2e":22.3,"variants":[{"variant":"","index":29.0,"costPerTask":0.49,"tokPerSec":119.0,"ttft":1.28,"e2e":22.3}]},{"slug":"step-5-preview","name":"Step 5 Preview","creator":"StepFun","context":1000000,"released":null,"index":29.0,"bestVariant":null,"costPerTask":2.35,"tokPerSec":165.0,"ttft":0.92,"e2e":16.03,"variants":[{"variant":"","index":29.0,"costPerTask":2.35,"tokPerSec":165.0,"ttft":0.92,"e2e":16.03}]},{"slug":"glm-5-3-flash","name":"GLM-5.3-Flash","creator":"Z AI","context":1000000,"released":null,"index":28.0,"bestVariant":null,"costPerTask":0.58,"tokPerSec":69.0,"ttft":1.48,"e2e":37.88,"variants":[{"variant":"","index":28.0,"costPerTask":0.58,"tokPerSec":69.0,"ttft":1.48,"e2e":37.88}]},{"slug":"gpt-6-luna","name":"GPT-6 Luna","creator":"OpenAI","context":1000000,"released":null,"index":28.0,"bestVariant":"max","costPerTask":0.15,"tokPerSec":118.0,"ttft":97.69,"e2e":101.92,"variants":[{"variant":"max","index":28.0,"costPerTask":0.15,"tokPerSec":118.0,"ttft":97.69,"e2e":101.92},{"variant":"xhigh","index":23.0,"costPerTask":0.1,"tokPerSec":119.0,"ttft":27.38,"e2e":31.58},{"variant":"high","index":21.0,"costPerTask":0.07,"tokPerSec":113.0,"ttft":12.27,"e2e":16.7},{"variant":"medium","index":18.0,"costPerTask":0.04,"tokPerSec":null,"ttft":null,"e2e":null},{"variant":"low","index":10.0,"costPerTask":0.01,"tokPerSec":125.0,"ttft":2.09,"e2e":6.1},{"variant":"non-reasoning","index":9.0,"costPerTask":0.07,"tokPerSec":115.0,"ttft":0.89,"e2e":5.24}]},{"slug":"mimo-v2-6-flash","name":"MiMo-V2.6-Flash","creator":"Xiaomi","context":1000000,"released":"2026-09-22","index":27.0,"bestVariant":null,"costPerTask":0.13,"tokPerSec":45.0,"ttft":3.1,"e2e":58.35,"variants":[{"variant":"","index":27.0,"costPerTask":0.13,"tokPerSec":45.0,"ttft":3.1,"e2e":58.35}]},{"slug":"mistral-large-4","name":"Mistral Large 4","creator":"Mistral","context":1000000,"released":null,"index":26.0,"bestVariant":"high","costPerTask":2.06,"tokPerSec":161.0,"ttft":1.42,"e2e":16.92,"variants":[{"variant":"high","index":26.0,"costPerTask":2.06,"tokPerSec":161.0,"ttft":1.42,"e2e":16.92},{"variant":"medium","index":23.0,"costPerTask":1.41,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"muse-spark-open","name":"Muse Spark Open","creator":"Meta","context":1050000,"released":null,"index":25.0,"bestVariant":"xhigh","costPerTask":2.02,"tokPerSec":139.0,"ttft":0.73,"e2e":18.67,"variants":[{"variant":"xhigh","index":25.0,"costPerTask":2.02,"tokPerSec":139.0,"ttft":0.73,"e2e":18.67}]},{"slug":"deepseek-v4-pro-0813","name":"DeepSeek V4 Pro 0813","creator":"DeepSeek","context":1000000,"released":null,"index":23.0,"bestVariant":"max","costPerTask":1.77,"tokPerSec":81.0,"ttft":1.66,"e2e":32.63,"variants":[{"variant":"max","index":23.0,"costPerTask":1.77,"tokPerSec":81.0,"ttft":1.66,"e2e":32.63},{"variant":"non-reasoning","index":null,"costPerTask":null,"tokPerSec":87.0,"ttft":1.72,"e2e":7.44}]},{"slug":"qwen3-8-27b","name":"Qwen3.8 27B","creator":"Alibaba","context":256000,"released":null,"index":20.0,"bestVariant":"xhigh","costPerTask":3.53,"tokPerSec":38.0,"ttft":4.06,"e2e":69.31,"variants":[{"variant":"xhigh","index":20.0,"costPerTask":3.53,"tokPerSec":38.0,"ttft":4.06,"e2e":69.31},{"variant":"medium","index":17.0,"costPerTask":2.16,"tokPerSec":42.0,"ttft":4.1,"e2e":62.99},{"variant":"low","index":16.0,"costPerTask":1.91,"tokPerSec":43.0,"ttft":3.82,"e2e":62.02},{"variant":"non-reasoning","index":10.0,"costPerTask":3.77,"tokPerSec":43.0,"ttft":3.99,"e2e":15.64}]},{"slug":"minimax-m3","name":"MiniMax-M3","creator":"MiniMax","context":1000000,"released":null,"index":19.0,"bestVariant":null,"costPerTask":1.29,"tokPerSec":106.0,"ttft":1.11,"e2e":24.64,"variants":[{"variant":"","index":19.0,"costPerTask":1.29,"tokPerSec":106.0,"ttft":1.11,"e2e":24.64}]},{"slug":"beam-1-0","name":"Beam 1.0","creator":"Reflection","context":1050000,"released":null,"index":18.0,"bestVariant":null,"costPerTask":0.28,"tokPerSec":299.0,"ttft":0.64,"e2e":9.0,"variants":[{"variant":"","index":18.0,"costPerTask":0.28,"tokPerSec":299.0,"ttft":0.64,"e2e":9.0}]},{"slug":"k2-horizon-375b-a23b","name":"K2 Horizon 375B A23B","creator":"Institute of Foundation Models","context":524000,"released":null,"index":18.0,"bestVariant":null,"costPerTask":0.0,"tokPerSec":154.0,"ttft":2.33,"e2e":18.55,"variants":[{"variant":"","index":18.0,"costPerTask":0.0,"tokPerSec":154.0,"ttft":2.33,"e2e":18.55}]},{"slug":"nemotron-3-5-super-vl","name":"Nemotron 3.5 Super VL","creator":"NVIDIA","context":1000000,"released":null,"index":14.0,"bestVariant":null,"costPerTask":0.54,"tokPerSec":474.0,"ttft":0.71,"e2e":5.99,"variants":[{"variant":"","index":14.0,"costPerTask":0.54,"tokPerSec":474.0,"ttft":0.71,"e2e":5.99}]},{"slug":"inkling-small","name":"Inkling Small","creator":"Thinking Machines","context":1000000,"released":null,"index":13.0,"bestVariant":null,"costPerTask":0.2,"tokPerSec":204.0,"ttft":1.54,"e2e":13.79,"variants":[{"variant":"","index":13.0,"costPerTask":0.2,"tokPerSec":204.0,"ttft":1.54,"e2e":13.79}]},{"slug":"gemini-3-5-flash-lite","name":"Gemini 3.5 Flash-Lite","creator":"Google","context":1000000,"released":null,"index":12.0,"bestVariant":null,"costPerTask":0.69,"tokPerSec":406.0,"ttft":6.59,"e2e":7.82,"variants":[{"variant":"","index":12.0,"costPerTask":0.69,"tokPerSec":406.0,"ttft":6.59,"e2e":7.82}]},{"slug":"inkling","name":"Inkling","creator":"Thinking Machines","context":1000000,"released":null,"index":12.0,"bestVariant":"xhigh","costPerTask":1.35,"tokPerSec":196.0,"ttft":1.81,"e2e":14.58,"variants":[{"variant":"xhigh","index":12.0,"costPerTask":1.35,"tokPerSec":196.0,"ttft":1.81,"e2e":14.58}]},{"slug":"nemotron-3-ultra","name":"Nemotron 3 Ultra","creator":"NVIDIA","context":262000,"released":null,"index":12.0,"bestVariant":null,"costPerTask":2.23,"tokPerSec":153.0,"ttft":1.29,"e2e":19.45,"variants":[{"variant":"","index":12.0,"costPerTask":2.23,"tokPerSec":153.0,"ttft":1.29,"e2e":19.45}]},{"slug":"command-a-plus","name":"Command A+","creator":"Cohere","context":192000,"released":null,"index":6.0,"bestVariant":null,"costPerTask":0.0,"tokPerSec":148.0,"ttft":0.5,"e2e":17.38,"variants":[{"variant":"","index":6.0,"costPerTask":0.0,"tokPerSec":148.0,"ttft":0.5,"e2e":17.38}]},{"slug":"muse-glimmer","name":"Muse Glimmer","creator":"Meta","context":131000,"released":null,"index":6.0,"bestVariant":"high","costPerTask":0.16,"tokPerSec":106.0,"ttft":1.21,"e2e":24.73,"variants":[{"variant":"high","index":6.0,"costPerTask":0.16,"tokPerSec":106.0,"ttft":1.21,"e2e":24.73}]},{"slug":"nemotron-3-5-lightning","name":"Nemotron 3.5 Lightning","creator":"NVIDIA","context":1000000,"released":null,"index":6.0,"bestVariant":null,"costPerTask":0.29,"tokPerSec":293.0,"ttft":0.56,"e2e":9.09,"variants":[{"variant":"","index":6.0,"costPerTask":0.29,"tokPerSec":293.0,"ttft":0.56,"e2e":9.09}]},{"slug":"granite-4-2-3b","name":"Granite 4.2 3B","creator":"IBM","context":131000,"released":null,"index":5.0,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":5.0,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"mistral-small-4","name":"Mistral Small 4","creator":"Mistral","context":256000,"released":null,"index":4.0,"bestVariant":null,"costPerTask":0.04,"tokPerSec":172.0,"ttft":0.83,"e2e":15.4,"variants":[{"variant":"","index":4.0,"costPerTask":0.04,"tokPerSec":172.0,"ttft":0.83,"e2e":15.4},{"variant":"non-reasoning","index":null,"costPerTask":null,"tokPerSec":162.0,"ttft":0.88,"e2e":3.96}]},{"slug":"nemotron-3-nano-4b","name":"Nemotron 3 Nano 4B","creator":"NVIDIA","context":262000,"released":null,"index":2.0,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":2.0,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"a-x-k2","name":"A.X-K2","creator":"SK Telecom","context":262000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"agnes-2-5-pro-beta","name":"Agnes 2.5 Pro Beta","creator":"Sapiens AI","context":1000000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":120.0,"ttft":1.35,"e2e":22.12,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":120.0,"ttft":1.35,"e2e":22.12}]},{"slug":"apertus-70b-instruct","name":"Apertus 70B Instruct","creator":"Swiss AI Initiative","context":65500,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"apertus-8b-instruct","name":"Apertus 8B Instruct","creator":"Swiss AI Initiative","context":65500,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":148.0,"ttft":3.31,"e2e":6.69,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":148.0,"ttft":3.31,"e2e":6.69}]},{"slug":"apodex-1-1","name":"Apodex 1.1","creator":"Apodex","context":256000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"apriel-v1-6-15b-thinker","name":"Apriel-v1.6-15B-Thinker","creator":"ServiceNow","context":128000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"celeris-1","name":"Celeris-1","creator":"Celeris","context":131000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":1680.0,"ttft":0.56,"e2e":0.86,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":1680.0,"ttft":0.56,"e2e":0.86}]},{"slug":"cogito-v2-1","name":"Cogito v2.1","creator":"Deep Cogito","context":128000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"command-a","name":"Command A","creator":"Cohere","context":256000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":18.0,"ttft":1.87,"e2e":29.87,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":18.0,"ttft":1.87,"e2e":29.87}]},{"slug":"deephermes-3-llama-3-1-8b","name":"DeepHermes 3 - Llama-3.1 8B","creator":"Nous Research","context":128000,"released":null,"index":null,"bestVariant":"non-reasoning","costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"non-reasoning","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"deephermes-3-mistral-24b","name":"DeepHermes 3 - Mistral 24B","creator":"Nous Research","context":32000,"released":null,"index":null,"bestVariant":"non-reasoning","costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"non-reasoning","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"deepseek-v4-flash-vision","name":"DeepSeek V4 Flash Vision","creator":"DeepSeek","context":1000000,"released":null,"index":null,"bestVariant":"max","costPerTask":null,"tokPerSec":163.0,"ttft":1.41,"e2e":16.78,"variants":[{"variant":"max","index":null,"costPerTask":null,"tokPerSec":163.0,"ttft":1.41,"e2e":16.78}]},{"slug":"diffusiongemma-26b-a4b","name":"DiffusionGemma 26B A4B","creator":"Google","context":256000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"doubao-seed-code","name":"Doubao Seed Code","creator":"ByteDance Seed","context":256000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"ernie-4-5-300b-a47b","name":"ERNIE 4.5 300B A47B","creator":"Baidu","context":131000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"ernie-5-0-thinking-preview","name":"ERNIE 5.0 Thinking Preview","creator":"Baidu","context":128000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"exaone-4-0-32b","name":"EXAONE 4.0 32B","creator":"LG AI Research","context":131000,"released":null,"index":null,"bestVariant":"non-reasoning","costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"non-reasoning","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null},{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"exaone-4-5-33b","name":"EXAONE 4.5 33B","creator":"LG AI Research","context":262000,"released":null,"index":null,"bestVariant":"non-reasoning","costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"non-reasoning","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null},{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"exaone-4-0-1-2b","name":"Exaone 4.0 1.2B","creator":"LG AI Research","context":64000,"released":null,"index":null,"bestVariant":"non-reasoning","costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"non-reasoning","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null},{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"falcon-h1r-7b","name":"Falcon-H1R-7B","creator":"TII UAE","context":256000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"g9v3-39a5b","name":"G9v3-39A5B","creator":"AI9Stars","context":131000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"g9v3-3b","name":"G9v3-3B","creator":"AI9Stars","context":131000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"gpt-5-3-codex","name":"GPT-5.3 Codex","creator":"OpenAI","context":400000,"released":null,"index":null,"bestVariant":"xhigh","costPerTask":null,"tokPerSec":79.0,"ttft":66.26,"e2e":72.57,"variants":[{"variant":"xhigh","index":null,"costPerTask":null,"tokPerSec":79.0,"ttft":66.26,"e2e":72.57}]},{"slug":"gpt-5-5-instant","name":"GPT-5.5 Instant","creator":"OpenAI","context":400000,"released":null,"index":null,"bestVariant":"June 2026","costPerTask":null,"tokPerSec":145.0,"ttft":0.92,"e2e":18.2,"variants":[{"variant":"June 2026","index":null,"costPerTask":null,"tokPerSec":145.0,"ttft":0.92,"e2e":18.2}]},{"slug":"gpt-5-5-pro","name":"GPT-5.5 Pro","creator":"OpenAI","context":922000,"released":null,"index":null,"bestVariant":"xhigh","costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"xhigh","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"gpt-6-sol","name":"GPT-6 Sol","creator":"OpenAI","context":1050000,"released":null,"index":null,"bestVariant":"Daybreak Blue, max","costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"Daybreak Blue, max","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"gemini-3-deep-think","name":"Gemini 3 Deep Think","creator":"Google","context":128000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"gemini-3-1-pro-preview","name":"Gemini 3.1 Pro Preview","creator":"Google","context":1000000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":134.0,"ttft":21.82,"e2e":25.55,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":134.0,"ttft":21.82,"e2e":25.55}]},{"slug":"gemma-3-270m","name":"Gemma 3 270M","creator":"Google","context":32000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"gemma-4-12b","name":"Gemma 4 12B","creator":"Google","context":262000,"released":null,"index":null,"bestVariant":"non-reasoning","costPerTask":null,"tokPerSec":118.0,"ttft":2.28,"e2e":6.52,"variants":[{"variant":"non-reasoning","index":null,"costPerTask":null,"tokPerSec":118.0,"ttft":2.28,"e2e":6.52},{"variant":"","index":null,"costPerTask":null,"tokPerSec":113.0,"ttft":2.22,"e2e":24.29}]},{"slug":"gemma-4-26b-a4b","name":"Gemma 4 26B A4B","creator":"Google","context":256000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":100.0,"ttft":0.95,"e2e":25.97,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":100.0,"ttft":0.95,"e2e":25.97},{"variant":"non-reasoning","index":null,"costPerTask":null,"tokPerSec":119.0,"ttft":0.97,"e2e":5.16}]},{"slug":"gemma-4-31b","name":"Gemma 4 31B","creator":"Google","context":256000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":35.0,"ttft":1.09,"e2e":64.74,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":35.0,"ttft":1.09,"e2e":64.74},{"variant":"non-reasoning","index":null,"costPerTask":null,"tokPerSec":61.0,"ttft":1.77,"e2e":9.99}]},{"slug":"gemma-4-e2b","name":"Gemma 4 E2B","creator":"Google","context":128000,"released":null,"index":null,"bestVariant":"non-reasoning","costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"non-reasoning","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null},{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"gemma-4-e4b","name":"Gemma 4 E4B","creator":"Google","context":128000,"released":null,"index":null,"bestVariant":"non-reasoning","costPerTask":null,"tokPerSec":44.0,"ttft":0.87,"e2e":12.18,"variants":[{"variant":"non-reasoning","index":null,"costPerTask":null,"tokPerSec":44.0,"ttft":0.87,"e2e":12.18},{"variant":"","index":null,"costPerTask":null,"tokPerSec":43.0,"ttft":0.98,"e2e":59.26}]},{"slug":"granite-4-0-350m","name":"Granite 4.0 350M","creator":"IBM","context":32800,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"granite-4-0-h-1b","name":"Granite 4.0 H 1B","creator":"IBM","context":128000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"granite-4-0-h-350m","name":"Granite 4.0 H 350M","creator":"IBM","context":32800,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"granite-4-0-h-small","name":"Granite 4.0 H Small","creator":"IBM","context":128000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"granite-4-0-micro","name":"Granite 4.0 Micro","creator":"IBM","context":128000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"granite-4-2-30b","name":"Granite 4.2 30B","creator":"IBM","context":131000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"granite-4-2-8b","name":"Granite 4.2 8B","creator":"IBM","context":131000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":57.0,"ttft":1.05,"e2e":44.64,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":57.0,"ttft":1.05,"e2e":44.64}]},{"slug":"hermes-4-405b","name":"Hermes 4 405B","creator":"Nous Research","context":128000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":40.0,"ttft":2.38,"e2e":65.43,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":40.0,"ttft":2.38,"e2e":65.43},{"variant":"non-reasoning","index":null,"costPerTask":null,"tokPerSec":38.0,"ttft":2.34,"e2e":15.37}]},{"slug":"hermes-4-70b","name":"Hermes 4 70B","creator":"Nous Research","context":128000,"released":null,"index":null,"bestVariant":"non-reasoning","costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"non-reasoning","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null},{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"hy3","name":"Hy3","creator":"Tencent","context":256000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":86.0,"ttft":2.77,"e2e":31.81,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":86.0,"ttft":2.77,"e2e":31.81}]},{"slug":"hyperclova-x-seed-think","name":"HyperCLOVA X SEED Think","creator":"Naver","context":128000,"released":null,"index":null,"bestVariant":"32B","costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"32B","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"hypernova-60b-2605","name":"HyperNova 60B 2605","creator":"Multiverse Computing","context":131000,"released":null,"index":null,"bestVariant":"high","costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"high","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"intellect-3","name":"INTELLECT-3","creator":"Prime Intellect","context":131000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"jt-35b-flash","name":"JT-35B-Flash","creator":"China Mobile","context":256000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"jt-4-1-flash-236b-a21b","name":"JT-4.1 Flash 236B A21B","creator":"China Mobile","context":262000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null},{"variant":"non-reasoning","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"jt-mini","name":"JT-MINI","creator":"China Mobile","context":128000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"jamba-1-7-large","name":"Jamba 1.7 Large","creator":"AI21 Labs","context":256000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"jamba-1-7-mini","name":"Jamba 1.7 Mini","creator":"AI21 Labs","context":258000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"jamba-reasoning-3b","name":"Jamba Reasoning 3B","creator":"AI21 Labs","context":262000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"k-exaone-2-0","name":"K-EXAONE 2.0","creator":"LG AI Research","context":262000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"k2-horizon-0-9b","name":"K2 Horizon 0.9B","creator":"Institute of Foundation Models","context":131000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"k2-horizon-3-7b","name":"K2 Horizon 3.7B","creator":"Institute of Foundation Models","context":524000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"k2-horizon-7b","name":"K2 Horizon 7B","creator":"Institute of Foundation Models","context":524000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"k2-horizon-mova-36b-a4b","name":"K2 Horizon MoVA 36B A4B","creator":"Institute of Foundation Models","context":524000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"k2-think-v2","name":"K2 Think V2","creator":"Institute of Foundation Models","context":262000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"k2-v2","name":"K2-V2","creator":"Institute of Foundation Models","context":512000,"released":null,"index":null,"bestVariant":"high","costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"high","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null},{"variant":"low","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null},{"variant":"medium","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"kat-coder-pro-v2","name":"KAT-Coder-Pro V2","creator":"KwaiKAT","context":256000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"kimi-k2-7-code","name":"Kimi K2.7 Code","creator":"Kimi","context":256000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":47.0,"ttft":2.81,"e2e":61.29,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":47.0,"ttft":2.81,"e2e":61.29}]},{"slug":"kimi-linear-48b-a3b-instruct","name":"Kimi Linear 48B A3B Instruct","creator":"Kimi","context":1000000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"lfm2-24b-a2b","name":"LFM2 24B A2B","creator":"Liquid AI","context":32800,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"lfm2-5-1-2b-instruct","name":"LFM2.5-1.2B-Instruct","creator":"Liquid AI","context":32000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"lfm2-5-1-2b-thinking","name":"LFM2.5-1.2B-Thinking","creator":"Liquid AI","context":32000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"lfm2-5-2-6b","name":"LFM2.5-2.6B","creator":"Liquid AI","context":128000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"lfm2-5-8b-a1b","name":"LFM2.5-8B-A1B","creator":"Liquid AI","context":32800,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"lfm2-5-vl-1-6b","name":"LFM2.5-VL-1.6B","creator":"Liquid AI","context":32000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"ling-3-0-tiny","name":"Ling 3.0 Tiny","creator":"InclusionAI","context":262000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":56.0,"ttft":3.09,"e2e":47.38,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":56.0,"ttft":3.09,"e2e":47.38}]},{"slug":"ling-3-1-flash","name":"Ling 3.1 Flash","creator":"InclusionAI","context":1000000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":213.0,"ttft":1.9,"e2e":13.65,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":213.0,"ttft":1.9,"e2e":13.65}]},{"slug":"ling-3-0-flash-fin","name":"Ling-3.0-flash-Fin","creator":"InclusionAI","context":262000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"ling-3-0-flash-vl","name":"Ling-3.0-flash-VL","creator":"InclusionAI","context":262000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"llama-3-1-405b","name":"Llama 3.1 405B","creator":"Meta","context":128000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"llama-3-2-11b","name":"Llama 3.2 11B","creator":"Meta","context":128000,"released":null,"index":null,"bestVariant":"Vision","costPerTask":null,"tokPerSec":30.0,"ttft":1.28,"e2e":17.76,"variants":[{"variant":"Vision","index":null,"costPerTask":null,"tokPerSec":30.0,"ttft":1.28,"e2e":17.76}]},{"slug":"llama-3-2-90b","name":"Llama 3.2 90B","creator":"Meta","context":128000,"released":null,"index":null,"bestVariant":"Vision","costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"Vision","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"llama-3-3-70b","name":"Llama 3.3 70B","creator":"Meta","context":128000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":91.0,"ttft":1.86,"e2e":7.34,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":91.0,"ttft":1.86,"e2e":7.34}]},{"slug":"llama-4-maverick","name":"Llama 4 Maverick","creator":"Meta","context":1000000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":48.0,"ttft":0.97,"e2e":11.36,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":48.0,"ttft":0.97,"e2e":11.36}]},{"slug":"llama-4-scout","name":"Llama 4 Scout","creator":"Meta","context":10000000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":133.0,"ttft":0.82,"e2e":4.59,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":133.0,"ttft":0.82,"e2e":4.59}]},{"slug":"llama-nemotron-ultra","name":"Llama Nemotron Ultra","creator":"NVIDIA","context":128000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"longcat-2-0","name":"LongCat 2.0","creator":"LongCat","context":1000000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":77.0,"ttft":2.64,"e2e":35.0,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":77.0,"ttft":2.64,"e2e":35.0}]},{"slug":"longcat-flash-lite","name":"LongCat Flash Lite","creator":"LongCat","context":256000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"magistral-medium-1-2","name":"Magistral Medium 1.2","creator":"Mistral","context":128000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"magistral-small-1-2","name":"Magistral Small 1.2","creator":"Mistral","context":128000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":106.0,"ttft":1.12,"e2e":24.73,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":106.0,"ttft":1.12,"e2e":24.73}]},{"slug":"mercury-2-5","name":"Mercury 2.5","creator":"Inception","context":260000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"mi-dm-k-2-5-pro","name":"Mi:dm K 2.5 Pro","creator":"Korea Telecom","context":128000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"minicpm-v-4-6-1-3b","name":"MiniCPM-V 4.6 1.3B","creator":"OpenBMB","context":262000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"minicpm5-1b","name":"MiniCPM5-1B","creator":"OpenBMB","context":128000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null},{"variant":"non-reasoning","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"minicpm5-2b","name":"MiniCPM5-2B","creator":"OpenBMB","context":131000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"ministral-3-14b","name":"Ministral 3 14B","creator":"Mistral","context":256000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":57.0,"ttft":1.05,"e2e":9.87,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":57.0,"ttft":1.05,"e2e":9.87}]},{"slug":"ministral-3-3b","name":"Ministral 3 3B","creator":"Mistral","context":256000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":181.0,"ttft":0.86,"e2e":3.62,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":181.0,"ttft":0.86,"e2e":3.62}]},{"slug":"ministral-3-8b","name":"Ministral 3 8B","creator":"Mistral","context":256000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":91.0,"ttft":0.84,"e2e":6.33,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":91.0,"ttft":0.84,"e2e":6.33}]},{"slug":"mistral-medium-3-5","name":"Mistral Medium 3.5","creator":"Mistral","context":256000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":157.0,"ttft":2.33,"e2e":18.28,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":157.0,"ttft":2.33,"e2e":18.28}]},{"slug":"molmo-7b-d","name":"Molmo 7B-D","creator":"Allen Institute for AI","context":4100,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"molmo2-8b","name":"Molmo2-8B","creator":"Allen Institute for AI","context":36900,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"motif-3","name":"Motif 3","creator":"Motif Technologies","context":262000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"motif-2-12-7b","name":"Motif-2-12.7B","creator":"Motif Technologies","context":128000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"nvidia-nemotron-nano-12b-v2-vl","name":"NVIDIA Nemotron Nano 12B v2 VL","creator":"NVIDIA","context":128000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null},{"variant":"non-reasoning","index":null,"costPerTask":null,"tokPerSec":203.0,"ttft":1.06,"e2e":3.52}]},{"slug":"nvidia-nemotron-nano-9b-v2","name":"NVIDIA Nemotron Nano 9B V2","creator":"NVIDIA","context":131000,"released":null,"index":null,"bestVariant":"non-reasoning","costPerTask":null,"tokPerSec":172.0,"ttft":0.97,"e2e":3.88,"variants":[{"variant":"non-reasoning","index":null,"costPerTask":null,"tokPerSec":172.0,"ttft":0.97,"e2e":3.88},{"variant":"","index":null,"costPerTask":null,"tokPerSec":213.0,"ttft":0.53,"e2e":12.28}]},{"slug":"nanbeige4-1-3b","name":"Nanbeige4.1-3B","creator":"Nanbeige","context":256000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"nemotron-3-nano","name":"Nemotron 3 Nano","creator":"NVIDIA","context":1000000,"released":null,"index":null,"bestVariant":"non-reasoning","costPerTask":null,"tokPerSec":243.0,"ttft":0.55,"e2e":2.6,"variants":[{"variant":"non-reasoning","index":null,"costPerTask":null,"tokPerSec":243.0,"ttft":0.55,"e2e":2.6},{"variant":"","index":null,"costPerTask":null,"tokPerSec":255.0,"ttft":0.93,"e2e":10.75}]},{"slug":"nemotron-3-nano-omni-30b-a3b","name":"Nemotron 3 Nano Omni 30B A3B","creator":"NVIDIA","context":256000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":231.0,"ttft":0.92,"e2e":11.76,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":231.0,"ttft":0.92,"e2e":11.76}]},{"slug":"nemotron-3-5-bolt-74b-a9b","name":"Nemotron 3.5 Bolt 74B-A9B","creator":"NVIDIA","context":262000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"nemotron-3-5-bolt-e4b-4gb","name":"Nemotron 3.5 Bolt E4B 4GB","creator":"NVIDIA","context":131000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"nemotron-cascade-2-30b-a3b","name":"Nemotron Cascade 2 30B A3B","creator":"NVIDIA","context":1000000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"nex-n2-pro","name":"Nex-N2-Pro","creator":"Nex AGI","context":262000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"north-mini-code","name":"North Mini Code","creator":"Cohere","context":256000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":70.0,"ttft":0.42,"e2e":36.39,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":70.0,"ttft":0.42,"e2e":36.39}]},{"slug":"nova-2-0-lite","name":"Nova 2.0 Lite","creator":"Amazon","context":1000000,"released":null,"index":null,"bestVariant":"high","costPerTask":null,"tokPerSec":183.0,"ttft":14.37,"e2e":28.0,"variants":[{"variant":"high","index":null,"costPerTask":null,"tokPerSec":183.0,"ttft":14.37,"e2e":28.0},{"variant":"low","index":null,"costPerTask":null,"tokPerSec":168.0,"ttft":9.98,"e2e":24.83},{"variant":"non-reasoning","index":null,"costPerTask":null,"tokPerSec":162.0,"ttft":1.12,"e2e":4.21},{"variant":"medium","index":null,"costPerTask":null,"tokPerSec":178.0,"ttft":14.65,"e2e":28.67}]},{"slug":"nova-2-0-omni","name":"Nova 2.0 Omni","creator":"Amazon","context":1000000,"released":null,"index":null,"bestVariant":"medium","costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"medium","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null},{"variant":"non-reasoning","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null},{"variant":"low","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"nova-2-0-pro-preview","name":"Nova 2.0 Pro Preview","creator":"Amazon","context":256000,"released":null,"index":null,"bestVariant":"medium","costPerTask":null,"tokPerSec":121.0,"ttft":11.4,"e2e":32.04,"variants":[{"variant":"medium","index":null,"costPerTask":null,"tokPerSec":121.0,"ttft":11.4,"e2e":32.04},{"variant":"non-reasoning","index":null,"costPerTask":null,"tokPerSec":119.0,"ttft":1.05,"e2e":5.25},{"variant":"low","index":null,"costPerTask":null,"tokPerSec":119.0,"ttft":8.64,"e2e":29.7}]},{"slug":"nova-micro","name":"Nova Micro","creator":"Amazon","context":130000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":272.0,"ttft":0.83,"e2e":2.67,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":272.0,"ttft":0.83,"e2e":2.67}]},{"slug":"nova-premier","name":"Nova Premier","creator":"Amazon","context":1000000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"olmo-3-7b","name":"Olmo 3 7B","creator":"Allen Institute for AI","context":65500,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"olmo-3-7b-think","name":"Olmo 3 7B Think","creator":"Allen Institute for AI","context":65500,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"olmo-3-1-32b-instruct","name":"Olmo 3.1 32B Instruct","creator":"Allen Institute for AI","context":65500,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"olmo-3-1-32b-think","name":"Olmo 3.1 32B Think","creator":"Allen Institute for AI","context":65500,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"phi-4","name":"Phi-4","creator":"Microsoft","context":16000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":41.0,"ttft":2.71,"e2e":15.02,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":41.0,"ttft":2.71,"e2e":15.02}]},{"slug":"phi-4-mini","name":"Phi-4 Mini","creator":"Microsoft","context":128000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":45.0,"ttft":0.87,"e2e":11.9,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":45.0,"ttft":0.87,"e2e":11.9}]},{"slug":"phi-4-multimodal","name":"Phi-4 Multimodal","creator":"Microsoft","context":128000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"quasar-438b","name":"Quasar 438B","creator":"Multiverse Computing","context":1000000,"released":null,"index":null,"bestVariant":"max","costPerTask":null,"tokPerSec":140.0,"ttft":1.37,"e2e":19.2,"variants":[{"variant":"max","index":null,"costPerTask":null,"tokPerSec":140.0,"ttft":1.37,"e2e":19.2}]},{"slug":"qwen3-coder-next","name":"Qwen3 Coder Next","creator":"Alibaba","context":256000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":78.0,"ttft":0.99,"e2e":7.42,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":78.0,"ttft":0.99,"e2e":7.42}]},{"slug":"qwen3-next-80b-a3b","name":"Qwen3 Next 80B A3B","creator":"Alibaba","context":262000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":130.0,"ttft":0.87,"e2e":4.7,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":130.0,"ttft":0.87,"e2e":4.7},{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"qwen3-omni-30b-a3b","name":"Qwen3 Omni 30B A3B","creator":"Alibaba","context":65500,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":96.0,"ttft":1.84,"e2e":7.07,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":96.0,"ttft":1.84,"e2e":7.07},{"variant":"","index":null,"costPerTask":null,"tokPerSec":102.0,"ttft":1.9,"e2e":26.37}]},{"slug":"qwen3-5-0-8b","name":"Qwen3.5 0.8B","creator":"Alibaba","context":262000,"released":null,"index":null,"bestVariant":"non-reasoning","costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"non-reasoning","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null},{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"qwen3-5-122b-a10b","name":"Qwen3.5 122B A10B","creator":"Alibaba","context":262000,"released":null,"index":null,"bestVariant":"non-reasoning","costPerTask":null,"tokPerSec":142.0,"ttft":2.28,"e2e":5.8,"variants":[{"variant":"non-reasoning","index":null,"costPerTask":null,"tokPerSec":142.0,"ttft":2.28,"e2e":5.8},{"variant":"","index":null,"costPerTask":null,"tokPerSec":129.0,"ttft":2.31,"e2e":21.68}]},{"slug":"qwen3-5-2b","name":"Qwen3.5 2B","creator":"Alibaba","context":262000,"released":null,"index":null,"bestVariant":"non-reasoning","costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"non-reasoning","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null},{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"qwen3-5-397b-a17b","name":"Qwen3.5 397B A17B","creator":"Alibaba","context":262000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":59.0,"ttft":2.24,"e2e":64.35,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":59.0,"ttft":2.24,"e2e":64.35},{"variant":"non-reasoning","index":null,"costPerTask":null,"tokPerSec":60.0,"ttft":2.15,"e2e":10.55}]},{"slug":"qwen3-5-4b","name":"Qwen3.5 4B","creator":"Alibaba","context":262000,"released":null,"index":null,"bestVariant":"non-reasoning","costPerTask":null,"tokPerSec":15.0,"ttft":0.99,"e2e":34.41,"variants":[{"variant":"non-reasoning","index":null,"costPerTask":null,"tokPerSec":15.0,"ttft":0.99,"e2e":34.41},{"variant":"","index":null,"costPerTask":null,"tokPerSec":13.0,"ttft":0.87,"e2e":187.59}]},{"slug":"qwen3-5-9b","name":"Qwen3.5 9B","creator":"Alibaba","context":262000,"released":null,"index":null,"bestVariant":"non-reasoning","costPerTask":null,"tokPerSec":88.0,"ttft":0.73,"e2e":6.43,"variants":[{"variant":"non-reasoning","index":null,"costPerTask":null,"tokPerSec":88.0,"ttft":0.73,"e2e":6.43},{"variant":"","index":null,"costPerTask":null,"tokPerSec":61.0,"ttft":1.68,"e2e":42.76}]},{"slug":"qwen3-5-omni-flash","name":"Qwen3.5 Omni Flash","creator":"Alibaba","context":256000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":238.0,"ttft":1.78,"e2e":3.88,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":238.0,"ttft":1.78,"e2e":3.88}]},{"slug":"qwen3-5-omni-plus","name":"Qwen3.5 Omni Plus","creator":"Alibaba","context":256000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":104.0,"ttft":2.09,"e2e":6.89,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":104.0,"ttft":2.09,"e2e":6.89}]},{"slug":"qwen3-6-35b-a3b","name":"Qwen3.6 35B A3B","creator":"Alibaba","context":262000,"released":null,"index":null,"bestVariant":"non-reasoning","costPerTask":null,"tokPerSec":143.0,"ttft":1.96,"e2e":5.46,"variants":[{"variant":"non-reasoning","index":null,"costPerTask":null,"tokPerSec":143.0,"ttft":1.96,"e2e":5.46},{"variant":"","index":null,"costPerTask":null,"tokPerSec":133.0,"ttft":1.96,"e2e":46.37}]},{"slug":"qwen3-7-plus","name":"Qwen3.7 Plus","creator":"Alibaba","context":1000000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":57.0,"ttft":2.25,"e2e":46.07,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":57.0,"ttft":2.25,"e2e":46.07}]},{"slug":"qwen3-8-2-4t-a95b","name":"Qwen3.8 2.4T A95B","creator":"Alibaba","context":262000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":39.0,"ttft":2.9,"e2e":66.72,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":39.0,"ttft":2.9,"e2e":66.72}]},{"slug":"qwen3-8-flash-next","name":"Qwen3.8-Flash-Next","creator":"Alibaba","context":256000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":51.0,"ttft":2.36,"e2e":51.29,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":51.0,"ttft":2.36,"e2e":51.29}]},{"slug":"r1-1776","name":"R1 1776","creator":"Perplexity","context":128000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"reka-flash-3","name":"Reka Flash 3","creator":"Reka AI","context":128000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":89.0,"ttft":1.38,"e2e":29.54,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":89.0,"ttft":1.38,"e2e":29.54}]},{"slug":"ring-2-6-1t","name":"Ring-2.6-1T","creator":"InclusionAI","context":262000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":114.0,"ttft":3.62,"e2e":25.62,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":114.0,"ttft":3.62,"e2e":25.62}]},{"slug":"ring-flash-2-0","name":"Ring-flash-2.0","creator":"InclusionAI","context":128000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"sarvam-105b","name":"Sarvam 105B","creator":"Sarvam","context":128000,"released":null,"index":null,"bestVariant":"high","costPerTask":null,"tokPerSec":159.0,"ttft":1.62,"e2e":17.38,"variants":[{"variant":"high","index":null,"costPerTask":null,"tokPerSec":159.0,"ttft":1.62,"e2e":17.38}]},{"slug":"sarvam-30b","name":"Sarvam 30B","creator":"Sarvam","context":65500,"released":null,"index":null,"bestVariant":"high","costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"high","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"solar-mini-4","name":"Solar Mini 4","creator":"Upstage","context":1050000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":118.0,"ttft":1.88,"e2e":23.06,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":118.0,"ttft":1.88,"e2e":23.06}]},{"slug":"solar-open-100b","name":"Solar Open 100B","creator":"Upstage","context":128000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"solar-open2-250b","name":"Solar Open2 250B","creator":"Upstage","context":1050000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"solar-pro-4","name":"Solar Pro 4","creator":"Upstage","context":512000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":99.0,"ttft":2.05,"e2e":27.22,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":99.0,"ttft":2.05,"e2e":27.22}]},{"slug":"step3-vl-10b","name":"Step3 VL 10B","creator":"StepFun","context":65500,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"tiny-aya-global","name":"Tiny Aya Global","creator":"Cohere","context":8189,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"tri-21b-think","name":"Tri-21B-Think","creator":"Trillion Labs","context":32000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"trinity-large-thinking","name":"Trinity Large Thinking","creator":"Arcee AI","context":512000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":null,"ttft":null,"e2e":null}]},{"slug":"gpt-oss-120b","name":"gpt-oss-120b","creator":"OpenAI","context":131000,"released":null,"index":null,"bestVariant":"low","costPerTask":null,"tokPerSec":250.0,"ttft":0.74,"e2e":10.73,"variants":[{"variant":"low","index":null,"costPerTask":null,"tokPerSec":250.0,"ttft":0.74,"e2e":10.73},{"variant":"high","index":null,"costPerTask":null,"tokPerSec":206.0,"ttft":0.81,"e2e":12.92}]},{"slug":"gpt-oss-20b","name":"gpt-oss-20b","creator":"OpenAI","context":131000,"released":null,"index":null,"bestVariant":"low","costPerTask":null,"tokPerSec":217.0,"ttft":0.82,"e2e":12.35,"variants":[{"variant":"low","index":null,"costPerTask":null,"tokPerSec":217.0,"ttft":0.82,"e2e":12.35},{"variant":"high","index":null,"costPerTask":null,"tokPerSec":143.0,"ttft":0.71,"e2e":18.19}]},{"slug":"o3","name":"o3","creator":"OpenAI","context":200000,"released":null,"index":null,"bestVariant":null,"costPerTask":null,"tokPerSec":147.0,"ttft":5.81,"e2e":9.21,"variants":[{"variant":"","index":null,"costPerTask":null,"tokPerSec":147.0,"ttft":5.81,"e2e":9.21}]}]},"legacy":{"asOf":"2026-03-08","models":[{"slug":"gpt-5-2","name":"GPT-5.2","category":"llm","provider":"OpenAI","type":"closed","license":"Prop","released":"2026.02","arch":"Dense · Reasoning","priceIn":1.75,"priceOut":14,"maxIn":400,"maxOut":128,"tokPerSec":null,"ttft":0.6,"modelUrl":"https://openai.com/gpt-5","scores":{"mmluPro":80,"gpqa":93.2,"aime":100,"hle":35.4,"arcAgi2":52.9,"metacog":62.76,"ifeval":90.5,"lcb":80,"sweV":80,"mmmlu":82,"termBench":64.9,"sciCode":54.6},"benchCount":12,"composite":73.02,"weighted":73.02,"confidence":{"cross-verified":7,"single-source":2,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"cross-verified","source":"OpenAI + Vellum + DataCamp"},"gpqa":{"level":"cross-verified","source":"OpenAI + Vellum + DataCamp"},"aime":{"level":"cross-verified","source":"OpenAI + Vellum + DataCamp"},"ifeval":{"level":"cross-verified","source":"OpenAI + Vellum + DataCamp"},"lcb":{"level":"cross-verified","source":"OpenAI + Vellum + DataCamp"},"sweV":{"level":"cross-verified","source":"OpenAI + Vellum + DataCamp"},"hle":{"level":"single-source","source":"OpenAI official"},"arcAgi2":{"level":"cross-verified","source":"OpenAI + arcprize.org"},"mmmlu":{"level":"single-source","source":"OpenAI official"}}},{"slug":"gpt-5-3-codex","name":"GPT-5.3 Codex","category":"llm","provider":"OpenAI","type":"closed","license":"Prop","released":"2026.02","arch":"Dense · Reasoning(Coding)","priceIn":7.5,"priceOut":30,"maxIn":400,"maxOut":128,"tokPerSec":null,"ttft":null,"modelUrl":"https://openai.com/codex","scores":{"mmluPro":82.9,"gpqa":91.5,"aime":95,"hle":36,"swePro":57,"mmmlu":83,"termBench":77.3},"benchCount":7,"composite":74.67,"weighted":62.47,"confidence":{"cross-verified":2,"single-source":0,"self-reported":0},"confidenceDetail":{"swePro":{"level":"cross-verified","source":"OpenAI + Scale AI SEAL"},"lcb":{"level":"cross-verified","source":"OpenAI + Scale AI SEAL"}}},{"slug":"claude-opus-4-6","name":"Claude Opus 4.6","category":"llm","provider":"Anthropic","type":"closed","license":"Prop","released":"2025.10","arch":"Dense · Reasoning(Adaptive)","priceIn":5,"priceOut":25,"maxIn":200,"maxOut":32,"tokPerSec":null,"ttft":3.5,"modelUrl":"https://anthropic.com/claude","scores":{"mmluPro":78.5,"gpqa":91.3,"aime":100,"hle":40,"arcAgi2":68.8,"metacog":56.04,"swePro":45,"ifeval":93.1,"lcb":76,"sweV":80.8,"mmmlu":80,"termBench":74.7},"benchCount":12,"composite":73.69,"weighted":73.69,"confidence":{"cross-verified":6,"single-source":3,"self-reported":0},"confidenceDetail":{"gpqa":{"level":"cross-verified","source":"Anthropic + Vellum + DataCamp"},"aime":{"level":"cross-verified","source":"Anthropic + Vellum + NxCode"},"hle":{"level":"cross-verified","source":"Vellum + digitalapplied"},"arcAgi2":{"level":"cross-verified","source":"Vellum + llm-stats + NxCode + DataCamp"},"swePro":{"level":"cross-verified","source":"Anthropic + Scale AI SEAL"},"ifeval":{"level":"cross-verified","source":"Anthropic + Vellum"},"lcb":{"level":"single-source","source":"Anthropic official"},"sweV":{"level":"single-source","source":"Anthropic official"},"metacog":{"level":"single-source","source":"FINAL Bench dataset"}}},{"slug":"gemini-3-1-pro","name":"Gemini 3.1 Pro","category":"llm","provider":"Google","type":"closed","license":"Prop","released":"2026.01","arch":"Dense · Reasoning(DeepThink)","priceIn":2,"priceOut":12,"maxIn":2000,"maxOut":64,"tokPerSec":null,"ttft":4.2,"modelUrl":"https://gemini.google.com","scores":{"mmluPro":83,"gpqa":94.3,"aime":97,"hle":44.4,"arcAgi2":77.1,"metacog":59.5,"swePro":43.3,"ifeval":91,"sweV":80.6,"mmmlu":88,"termBench":78.4,"sciCode":58.9},"benchCount":12,"composite":74.62,"weighted":74.62,"confidence":{"cross-verified":5,"single-source":5,"self-reported":0},"confidenceDetail":{"gpqa":{"level":"cross-verified","source":"Google DeepMind + PCMag + NxCode"},"aime":{"level":"cross-verified","source":"Google DeepMind + Vellum"},"hle":{"level":"cross-verified","source":"Google DeepMind + Vellum"},"arcAgi2":{"level":"cross-verified","source":"Google DeepMind + arcprize.org + NxCode + DataCamp"},"sweV":{"level":"cross-verified","source":"Google DeepMind + marc0.dev"},"mmluPro":{"level":"single-source","source":"Google DeepMind model card"},"swePro":{"level":"single-source","source":"Google DeepMind model card"},"ifeval":{"level":"single-source","source":"Google DeepMind model card"},"lcb":{"level":"single-source","source":"Google DeepMind model card"},"mmmlu":{"level":"single-source","source":"Google DeepMind model card"}}},{"slug":"grok-4-heavy","name":"Grok 4 Heavy","category":"llm","provider":"xAI","type":"closed","license":"Prop","released":"2025.11","arch":"Dense · Reasoning","priceIn":3,"priceOut":15,"maxIn":256,"maxOut":32,"tokPerSec":null,"ttft":6,"modelUrl":"https://x.ai/grok","scores":{"mmluPro":85,"gpqa":87.5,"aime":95,"hle":28,"swePro":25,"ifeval":88,"sweV":72,"mmmlu":76,"termBench":27.2},"benchCount":9,"composite":64.86,"weighted":61.53,"confidence":{"cross-verified":0,"single-source":8,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"xAI official"},"gpqa":{"level":"single-source","source":"xAI official"},"aime":{"level":"single-source","source":"xAI official"},"hle":{"level":"single-source","source":"xAI official"},"arcAgi2":{"level":"single-source","source":"xAI official"},"swePro":{"level":"single-source","source":"xAI official"},"ifeval":{"level":"single-source","source":"xAI official"},"lcb":{"level":"single-source","source":"xAI official"}}},{"slug":"claude-sonnet-4-6","name":"Claude Sonnet 4.6","category":"llm","provider":"Anthropic","type":"closed","license":"Prop","released":"2026.02","arch":"Dense · Reasoning","priceIn":3,"priceOut":15,"maxIn":200,"maxOut":64,"tokPerSec":null,"ttft":2,"modelUrl":"https://anthropic.com/claude","scores":{"gpqa":89.9,"aime":83,"arcAgi2":60.4,"ifeval":89.5,"sweV":79.6,"termBench":53},"benchCount":6,"composite":75.9,"weighted":58.79,"confidence":{"cross-verified":3,"single-source":3,"self-reported":0},"confidenceDetail":{"gpqa":{"level":"cross-verified","source":"Anthropic + Vellum + NxCode"},"aime":{"level":"cross-verified","source":"Anthropic + Vellum + NxCode"},"arcAgi2":{"level":"cross-verified","source":"Anthropic + Vellum + NxCode"},"swePro":{"level":"single-source","source":"Anthropic official"},"ifeval":{"level":"single-source","source":"Anthropic official"},"sweV":{"level":"single-source","source":"Anthropic official"}}},{"slug":"gpt-oss-120b","name":"GPT-OSS-120B","category":"llm","provider":"OpenAI","type":"open","license":"Apache2","released":"2025.12","arch":"MoE · Reasoning(5.1B/116.8B)","priceIn":0,"priceOut":0,"maxIn":128,"maxOut":32,"tokPerSec":null,"ttft":null,"modelUrl":"https://huggingface.co/openai","scores":{"mmluPro":90,"gpqa":80.9,"aime":97.9,"hle":12,"metacog":60.42,"swePro":16.2,"mmmlu":72},"benchCount":7,"composite":61.35,"weighted":51.33,"confidence":{"cross-verified":0,"single-source":3,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"OpenAI official"},"gpqa":{"level":"single-source","source":"OpenAI official"},"aime":{"level":"single-source","source":"OpenAI official"}}},{"slug":"gpt-oss-20b","name":"GPT-OSS-20B","category":"llm","provider":"OpenAI","type":"open","license":"Apache2","released":"2025.12","arch":"MoE · Reasoning(3.6B/20.9B)","priceIn":0,"priceOut":0,"maxIn":128,"maxOut":32,"tokPerSec":null,"ttft":null,"modelUrl":"https://huggingface.co/openai","scores":{"mmluPro":85.3,"gpqa":71.5,"aime":98.7,"hle":7,"mmmlu":68},"benchCount":5,"composite":66.1,"weighted":46.74,"confidence":{"cross-verified":0,"single-source":3,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"OpenAI official"},"gpqa":{"level":"single-source","source":"OpenAI official"},"aime":{"level":"single-source","source":"OpenAI official"}}},{"slug":"qwen3-5-397b","name":"Qwen3.5-397B","category":"llm","provider":"Alibaba","type":"open","license":"Apache2","released":"2026.01","arch":"MoE+Hybrid · Reasoning(17B/397B)","priceIn":null,"priceOut":null,"maxIn":262,"maxOut":32,"tokPerSec":45,"ttft":5,"modelUrl":"https://huggingface.co/Qwen","scores":{"mmluPro":87.8,"gpqa":88.4,"aime":91.3,"hle":32,"swePro":38,"ifeval":92.6,"sweV":76.4,"mmmlu":85},"benchCount":8,"composite":73.94,"weighted":66.13,"confidence":{"cross-verified":0,"single-source":6,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"Alibaba Cloud official"},"gpqa":{"level":"single-source","source":"Alibaba Cloud official"},"aime":{"level":"single-source","source":"Alibaba Cloud official"},"swePro":{"level":"single-source","source":"Alibaba Cloud official"},"ifeval":{"level":"single-source","source":"Alibaba Cloud official"},"lcb":{"level":"single-source","source":"Alibaba Cloud official"}}},{"slug":"qwen3-5-122b","name":"Qwen3.5-122B","category":"llm","provider":"Alibaba","type":"open","license":"Apache2","released":"2026.01","arch":"MoE+Hybrid · Reasoning(10B/122B)","priceIn":0.4,"priceOut":1.2,"maxIn":262,"maxOut":32,"tokPerSec":null,"ttft":6,"modelUrl":"https://huggingface.co/Qwen","scores":{"gpqa":87.6,"aime":85,"hle":24,"swePro":28,"bfcl":72.2,"ifeval":93.4,"sweV":72.2,"mmmlu":82},"benchCount":8,"composite":68.05,"weighted":60.87,"confidence":{"cross-verified":0,"single-source":4,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"Alibaba Cloud official"},"gpqa":{"level":"single-source","source":"Alibaba Cloud official"},"aime":{"level":"single-source","source":"Alibaba Cloud official"},"ifeval":{"level":"single-source","source":"Alibaba Cloud official"}}},{"slug":"qwen3-5-27b","name":"Qwen3.5-27B","category":"llm","provider":"Alibaba","type":"open","license":"Apache2","released":"2026.01","arch":"Dense · Reasoning(27B)","priceIn":null,"priceOut":null,"maxIn":262,"maxOut":32,"tokPerSec":null,"ttft":5.5,"modelUrl":"https://huggingface.co/Qwen","scores":{"mmluPro":86.1,"gpqa":85.5,"hle":18,"swePro":20,"sweV":72.4,"mmmlu":80},"benchCount":6,"composite":60.33,"weighted":46.73,"confidence":{"cross-verified":0,"single-source":3,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"Alibaba Cloud official"},"gpqa":{"level":"single-source","source":"Alibaba Cloud official"},"aime":{"level":"single-source","source":"Alibaba Cloud official"}}},{"slug":"qwen3-5-35b","name":"Qwen3.5-35B","category":"llm","provider":"Alibaba","type":"open","license":"Apache2","released":"2026.01","arch":"MoE+Hybrid · Reasoning(3B/35B)","priceIn":0.1,"priceOut":0.4,"maxIn":262,"maxOut":32,"tokPerSec":194,"ttft":4,"modelUrl":"https://huggingface.co/Qwen","scores":{"gpqa":83,"hle":15,"swePro":18,"sweV":68,"mmmlu":78},"benchCount":5,"composite":52.4,"weighted":37.05,"confidence":{"cross-verified":0,"single-source":3,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"Alibaba Cloud official"},"gpqa":{"level":"single-source","source":"Alibaba Cloud official"},"aime":{"level":"single-source","source":"Alibaba Cloud official"}}},{"slug":"qwen3-5-flash","name":"Qwen3.5-Flash","category":"llm","provider":"Alibaba","type":"closed","license":"Prop","released":"2026.01","arch":"MoE+Hybrid · Non-Reasoning","priceIn":0.1,"priceOut":0.4,"maxIn":1000,"maxOut":32,"tokPerSec":null,"ttft":3,"modelUrl":"https://huggingface.co/Qwen","scores":{"mmmlu":75},"benchCount":1,"composite":75.0,"weighted":23.72,"confidence":{"cross-verified":0,"single-source":3,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"Alibaba Cloud official"},"gpqa":{"level":"single-source","source":"Alibaba Cloud official"},"aime":{"level":"single-source","source":"Alibaba Cloud official"}}},{"slug":"qwen3-5-9b","name":"Qwen3.5-9B","category":"llm","provider":"Alibaba","type":"open","license":"Apache2","released":"2026.01","arch":"Dense · Reasoning(9B)","priceIn":null,"priceOut":null,"maxIn":262,"maxOut":32,"tokPerSec":null,"ttft":3.5,"modelUrl":"https://huggingface.co/Qwen","scores":{"mmluPro":82.5,"gpqa":81.7,"hle":10,"bfcl":66.1,"ifeval":91.5,"lcb":65.6,"mmmlu":81.2},"benchCount":7,"composite":68.37,"weighted":57.2,"confidence":{"cross-verified":0,"single-source":2,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"Alibaba Cloud official"},"gpqa":{"level":"single-source","source":"Alibaba Cloud official"}}},{"slug":"qwen3-5-4b","name":"Qwen3.5-4B","category":"llm","provider":"Alibaba","type":"open","license":"Apache2","released":"2026.01","arch":"Dense · Reasoning(4B)","priceIn":null,"priceOut":null,"maxIn":262,"maxOut":32,"tokPerSec":null,"ttft":2,"modelUrl":"https://huggingface.co/Qwen","scores":{"mmluPro":79.1,"gpqa":76.2,"bfcl":50.3,"ifeval":89.8,"lcb":55.8,"mmmlu":76.1},"benchCount":6,"composite":71.22,"weighted":55.16,"confidence":{"cross-verified":0,"single-source":2,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"Alibaba Cloud official"},"gpqa":{"level":"single-source","source":"Alibaba Cloud official"}}},{"slug":"deepseek-v3-2","name":"DeepSeek V3.2","category":"llm","provider":"DeepSeek","type":"open","license":"MIT","released":"2025.12","arch":"MoE · Non-Reasoning(37B/671B)","priceIn":0.14,"priceOut":0.28,"maxIn":128,"maxOut":8,"tokPerSec":null,"ttft":7,"modelUrl":"https://huggingface.co/deepseek-ai/DeepSeek-V3","scores":{"gpqa":79.9,"aime":89.3,"hle":22,"metacog":60.04,"swePro":15.6,"ifeval":87.5,"sweV":66,"mmmlu":78,"termBench":39.6},"benchCount":9,"composite":59.77,"weighted":56.7,"confidence":{"cross-verified":5,"single-source":0,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"cross-verified","source":"DeepSeek + AA Intelligence Index"},"gpqa":{"level":"cross-verified","source":"DeepSeek + AA Intelligence Index"},"aime":{"level":"cross-verified","source":"DeepSeek + AA Intelligence Index"},"ifeval":{"level":"cross-verified","source":"DeepSeek + AA Intelligence Index"},"lcb":{"level":"cross-verified","source":"DeepSeek + AA Intelligence Index"}}},{"slug":"deepseek-r1","name":"DeepSeek R1","category":"llm","provider":"DeepSeek","type":"open","license":"MIT","released":"2025.01","arch":"MoE · Reasoning(37B/671B)","priceIn":0.55,"priceOut":2.19,"maxIn":128,"maxOut":8,"tokPerSec":null,"ttft":8,"modelUrl":"https://huggingface.co/deepseek-ai/DeepSeek-R1","scores":{"mmluPro":85,"gpqa":82,"aime":87.5,"hle":14,"swePro":18,"ifeval":83.3,"sweV":49.2,"mmmlu":74},"benchCount":8,"composite":61.62,"weighted":55.12,"confidence":{"cross-verified":4,"single-source":0,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"cross-verified","source":"DeepSeek + AA Intelligence Index"},"gpqa":{"level":"cross-verified","source":"DeepSeek + AA Intelligence Index"},"aime":{"level":"cross-verified","source":"DeepSeek + AA Intelligence Index"},"ifeval":{"level":"cross-verified","source":"DeepSeek + AA Intelligence Index"}}},{"slug":"kimi-k2-5","name":"Kimi K2.5","category":"llm","provider":"Moonshot","type":"open","license":"MIT","released":"2025.11","arch":"MoE · Reasoning(1T)","priceIn":0.55,"priceOut":2.5,"maxIn":200,"maxOut":32,"tokPerSec":null,"ttft":5,"modelUrl":"https://huggingface.co/moonshotai","scores":{"mmluPro":92,"gpqa":87.6,"aime":96.1,"hle":44.9,"arcAgi2":12.1,"metacog":68.71,"swePro":27.7,"ifeval":94,"lcb":85,"sweV":76.8,"mmmlu":81,"termBench":43.2},"benchCount":12,"composite":67.43,"weighted":67.43,"confidence":{"cross-verified":4,"single-source":3,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"cross-verified","source":"Moonshot AI + AA Intelligence Index"},"gpqa":{"level":"cross-verified","source":"Moonshot AI + AA Intelligence Index"},"aime":{"level":"cross-verified","source":"Moonshot AI + AA Intelligence Index"},"hle":{"level":"cross-verified","source":"Moonshot AI + AA Intelligence Index"},"arcAgi2":{"level":"single-source","source":"Moonshot AI official"},"metacog":{"level":"single-source","source":"FINAL Bench dataset"},"lcb":{"level":"single-source","source":"Moonshot AI official"}}},{"slug":"glm-5-2","name":"GLM-5.2","category":"llm","provider":"Z.ai","type":"open","license":"MIT","released":"2026.06","arch":"MoE · Reasoning(40B/744B)","priceIn":null,"priceOut":null,"maxIn":1000,"maxOut":16,"tokPerSec":null,"ttft":null,"modelUrl":"https://huggingface.co/zai-org/GLM-5.2","scores":{"gpqa":91.2,"aime":99.2,"swePro":62.1},"benchCount":3,"composite":84.17,"weighted":46.1,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"glm-5","name":"GLM-5","category":"llm","provider":"Zhipu AI","type":"open","license":"MIT","released":"2026.02","arch":"MoE · Reasoning(40B/745B)","priceIn":0.35,"priceOut":0.39,"maxIn":200,"maxOut":16,"tokPerSec":null,"ttft":null,"modelUrl":"https://huggingface.co/zai-org/GLM-5","scores":{"mmluPro":87,"gpqa":86,"aime":92.7,"hle":30.5,"metacog":62.5,"swePro":77.8,"lcb":52,"mmmlu":76,"termBench":52.4},"benchCount":9,"composite":68.54,"weighted":65.03,"confidence":{"cross-verified":0,"single-source":5,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"Zhipu AI official"},"gpqa":{"level":"single-source","source":"Zhipu AI official"},"aime":{"level":"single-source","source":"Zhipu AI official"},"ifeval":{"level":"single-source","source":"Zhipu AI official"},"lcb":{"level":"single-source","source":"Zhipu AI official"}}},{"slug":"llama-4-scout","name":"Llama 4 Scout","category":"llm","provider":"Meta","type":"open","license":"Meta","released":"2025.04","arch":"MoE · Non-Reasoning(17B/400B)","priceIn":0.11,"priceOut":0.34,"maxIn":10000,"maxOut":16,"tokPerSec":null,"ttft":2,"modelUrl":"https://huggingface.co/meta-llama","scores":{"gpqa":73,"aime":85,"hle":12,"swePro":5.2,"ifeval":85,"sweV":55,"mmmlu":70},"benchCount":7,"composite":55.03,"weighted":46.04,"confidence":{"cross-verified":4,"single-source":0,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"cross-verified","source":"Meta + AA Intelligence Index"},"gpqa":{"level":"cross-verified","source":"Meta + AA Intelligence Index"},"aime":{"level":"cross-verified","source":"Meta + AA Intelligence Index"},"ifeval":{"level":"cross-verified","source":"Meta + AA Intelligence Index"}}},{"slug":"mistral-large-3","name":"Mistral Large 3","category":"llm","provider":"Mistral","type":"open","license":"Apache2","released":"2025.11","arch":"MoE · Non-Reasoning(675B)","priceIn":2,"priceOut":6,"maxIn":256,"maxOut":16,"tokPerSec":null,"ttft":0.3,"modelUrl":"https://huggingface.co/mistralai","scores":{"gpqa":78,"aime":82,"hle":11,"swePro":12,"ifeval":86,"sweV":60,"mmmlu":72},"benchCount":7,"composite":57.29,"weighted":47.93,"confidence":{"cross-verified":0,"single-source":3,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"Mistral AI official"},"gpqa":{"level":"single-source","source":"Mistral AI official"},"ifeval":{"level":"single-source","source":"Mistral AI official"}}},{"slug":"gemini-3-flash","name":"Gemini 3 Flash","category":"llm","provider":"Google","type":"closed","license":"Prop","released":"2025.12","arch":"Dense · Non-Reasoning+Thinking","priceIn":0.5,"priceOut":3,"maxIn":1000,"maxOut":64,"tokPerSec":218,"ttft":1.2,"modelUrl":"https://deepmind.google/technologies/gemini/flash/","scores":{"mmluPro":88.6,"gpqa":90.4,"aime":95,"hle":33.7,"arcAgi2":34,"swePro":71.2,"ifeval":88.2,"sweV":72.5,"mmmlu":83,"termBench":64.3},"benchCount":10,"composite":72.09,"weighted":72.09,"confidence":{"cross-verified":0,"single-source":7,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"Google DeepMind model card"},"gpqa":{"level":"single-source","source":"Google DeepMind model card"},"aime":{"level":"single-source","source":"Google DeepMind model card"},"ifeval":{"level":"single-source","source":"Google DeepMind model card"},"lcb":{"level":"single-source","source":"Google DeepMind model card"},"sweV":{"level":"single-source","source":"Google DeepMind model card"},"mmmlu":{"level":"single-source","source":"Google DeepMind model card"}}},{"slug":"llama-4-maverick","name":"Llama 4 Maverick","category":"llm","provider":"Meta","type":"open","license":"Llama4","released":"2025.10","arch":"MoE · Non-Reasoning(17B/400B)","priceIn":0.22,"priceOut":0.88,"maxIn":1000,"maxOut":16,"tokPerSec":null,"ttft":4.5,"modelUrl":"https://huggingface.co/meta-llama","scores":{"mmluPro":80.5,"gpqa":69.8,"aime":82,"hle":18,"swePro":12.3,"ifeval":83,"sweV":73,"mmmlu":74},"benchCount":8,"composite":61.58,"weighted":55.07,"confidence":{"cross-verified":4,"single-source":0,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"cross-verified","source":"Meta + AA Intelligence Index"},"gpqa":{"level":"cross-verified","source":"Meta + AA Intelligence Index"},"aime":{"level":"cross-verified","source":"Meta + AA Intelligence Index"},"ifeval":{"level":"cross-verified","source":"Meta + AA Intelligence Index"}}},{"slug":"claude-haiku-4-5","name":"Claude Haiku 4.5","category":"llm","provider":"Anthropic","type":"closed","license":"Prop","released":"2025.09","arch":"Dense · Non-Reasoning","priceIn":1,"priceOut":5,"maxIn":200,"maxOut":8,"tokPerSec":null,"ttft":0.4,"modelUrl":"https://anthropic.com/claude","scores":{"mmluPro":72,"gpqa":75,"swePro":14,"ifeval":86.5,"sweV":68,"mmmlu":71,"termBench":35.5},"benchCount":7,"composite":60.29,"weighted":50.44,"confidence":{"cross-verified":0,"single-source":4,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"Anthropic official"},"gpqa":{"level":"single-source","source":"Anthropic official"},"aime":{"level":"single-source","source":"Anthropic official"},"ifeval":{"level":"single-source","source":"Anthropic official"}}},{"slug":"grok-4-1-fast","name":"Grok 4.1 Fast","category":"llm","provider":"xAI","type":"closed","license":"Prop","released":"2025.11","arch":"Dense · Reasoning","priceIn":0.2,"priceOut":0.8,"maxIn":2000,"maxOut":16,"tokPerSec":null,"ttft":1.5,"modelUrl":"https://x.ai/grok","scores":{"gpqa":85.3,"aime":88,"mmmlu":72},"benchCount":3,"composite":81.77,"weighted":44.79,"confidence":{"cross-verified":0,"single-source":5,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"xAI official"},"gpqa":{"level":"single-source","source":"xAI official"},"aime":{"level":"single-source","source":"xAI official"},"swePro":{"level":"single-source","source":"xAI official"},"ifeval":{"level":"single-source","source":"xAI official"}}},{"slug":"deepseek-r2","name":"DeepSeek R2","category":"llm","provider":"DeepSeek","type":"open","license":"MIT","released":"2026.02","arch":"MoE · Reasoning(671B)","priceIn":0.55,"priceOut":2.19,"maxIn":128,"maxOut":8,"tokPerSec":null,"ttft":9,"modelUrl":"https://huggingface.co/deepseek-ai","scores":{"mmluPro":87,"gpqa":88,"aime":93.8,"ifeval":84,"mmmlu":76},"benchCount":5,"composite":85.76,"weighted":60.64,"confidence":{"cross-verified":0,"single-source":6,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"DeepSeek official"},"gpqa":{"level":"single-source","source":"DeepSeek official"},"aime":{"level":"single-source","source":"DeepSeek official"},"swePro":{"level":"single-source","source":"DeepSeek official"},"ifeval":{"level":"single-source","source":"DeepSeek official"},"lcb":{"level":"single-source","source":"DeepSeek official"}}},{"slug":"phi-4","name":"Phi-4","category":"llm","provider":"Microsoft","type":"open","license":"MIT","released":"2024.12","arch":"Dense · Non-Reasoning(14B)","priceIn":null,"priceOut":null,"maxIn":16,"maxOut":4,"tokPerSec":null,"ttft":2.5,"modelUrl":"https://huggingface.co/microsoft/phi-4","scores":{"gpqa":73,"sweV":72,"mmmlu":68},"benchCount":3,"composite":71.0,"weighted":38.89,"confidence":{"cross-verified":0,"single-source":3,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"Microsoft official"},"gpqa":{"level":"single-source","source":"Microsoft official"},"aime":{"level":"single-source","source":"Microsoft official"}}},{"slug":"k-exaone","name":"K-EXAONE","category":"llm","provider":"LG AI Research","type":"open","license":"Research","released":"2025.12","arch":"MoE · Reasoning(236B, K-Foundation)","priceIn":0,"priceOut":0,"maxIn":260,"maxOut":16,"tokPerSec":null,"ttft":null,"modelUrl":"https://huggingface.co/LGAI-EXAONE","scores":{"mmluPro":81.8,"gpqa":75.4,"aime":85.3,"sweV":49.4,"mmmlu":72},"benchCount":5,"composite":72.78,"weighted":51.46,"confidence":{"cross-verified":3,"single-source":1,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"cross-verified","source":"Korea Herald + K-EXAONE Technical Report"},"gpqa":{"level":"cross-verified","source":"Korea Herald + K-EXAONE Technical Report"},"aime":{"level":"cross-verified","source":"Korea Herald + K-EXAONE Technical Report"},"sweV":{"level":"single-source","source":"K-EXAONE Technical Report"}}},{"slug":"a-x-k1","name":"A.X K1","category":"llm","provider":"SK Telecom","type":"closed","license":"Apache2","released":"2025.12","arch":"MoE · Reasoning(500B, K-Foundation)","priceIn":0,"priceOut":0,"maxIn":64,"maxOut":16,"tokPerSec":null,"ttft":null,"modelUrl":"https://www.sktelecom.com","scores":{"mmmlu":62},"benchCount":1,"composite":62.0,"weighted":19.61,"confidence":{"cross-verified":0,"single-source":0,"self-reported":2},"confidenceDetail":{"mmluPro":{"level":"self-reported","source":"Self-reported benchmark"},"gpqa":{"level":"self-reported","source":"Self-reported benchmark"}}},{"slug":"solar-open-100b","name":"Solar Open 100B","category":"llm","provider":"Upstage","type":"open","license":"Apache2","released":"2025.12","arch":"MoE · Reasoning(100B, K-Foundation)","priceIn":0,"priceOut":0,"maxIn":100,"maxOut":16,"tokPerSec":null,"ttft":null,"modelUrl":"https://huggingface.co/upstage/Solar-Open-100B","scores":{"mmluPro":80.4,"gpqa":68.1,"aime":84.3,"swePro":74.2,"mmmlu":68},"benchCount":5,"composite":75.0,"weighted":53.03,"confidence":{"cross-verified":0,"single-source":0,"self-reported":2},"confidenceDetail":{"mmluPro":{"level":"self-reported","source":"Upstage self-reported"},"gpqa":{"level":"self-reported","source":"Upstage self-reported"}}},{"slug":"motif-ai","name":"Motif AI","category":"llm","provider":"Motif Technologies","type":"closed","license":"Prop","released":"2026.02","arch":"Dense · Non-Reasoning(K-Foundation)","priceIn":0,"priceOut":0,"maxIn":128,"maxOut":16,"tokPerSec":null,"ttft":null,"modelUrl":"https://motif.ai","scores":{"mmmlu":55},"benchCount":1,"composite":55.0,"weighted":17.39,"confidence":{"cross-verified":0,"single-source":0,"self-reported":3},"confidenceDetail":{"mmluPro":{"level":"self-reported","source":"Self-reported"},"gpqa":{"level":"self-reported","source":"Self-reported"},"aime":{"level":"self-reported","source":"Self-reported"}}},{"slug":"mi-dm-2-0-base","name":"Mi:dm 2.0 Base","category":"llm","provider":"KT","type":"open","license":"MIT","released":"2025.07","arch":"Dense · Non-Reasoning(11.5B, Llama-DuS)","priceIn":0,"priceOut":0,"maxIn":32,"maxOut":4,"tokPerSec":null,"ttft":null,"modelUrl":"https://huggingface.co/K-intelligence/Midm-2.0-Base-Instruct","scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":1},"confidenceDetail":{"mmluPro":{"level":"self-reported","source":"Self-reported"}}},{"slug":"nanbeige4-1-3b","name":"Nanbeige4.1-3B","category":"llm","provider":"Nanbeige","type":"open","license":"Apache2","released":"2026.03","arch":"Dense · Reasoning(3B)","priceIn":0,"priceOut":0,"maxIn":131,"maxOut":131,"tokPerSec":null,"ttft":null,"modelUrl":"https://huggingface.co/Nanbeige/Nanbeige4.1-3B","scores":{"gpqa":83.8,"hle":12.6,"bfcl":56.5,"lcb":76.9},"benchCount":4,"composite":57.45,"weighted":36.33,"confidence":{"cross-verified":0,"single-source":0,"self-reported":2},"confidenceDetail":{"mmluPro":{"level":"self-reported","source":"Self-reported"},"gpqa":{"level":"self-reported","source":"Self-reported"}}},{"slug":"minimax-m2-5","name":"MiniMax-M2.5","category":"llm","provider":"MiniMax","type":"open","license":"MIT","released":"2026.02","arch":"MoE · Reasoning(10B/230B)","priceIn":0.3,"priceOut":1.2,"maxIn":1000,"maxOut":64,"tokPerSec":47,"ttft":3.3,"modelUrl":"https://huggingface.co/MiniMaxAI/MiniMax-M2.5","scores":{"mmluPro":82,"gpqa":84.8,"aime":86.3,"hle":19.1,"metacog":60.54,"ifeval":87.5,"lcb":82.6,"sweV":80.2,"termBench":42.2},"benchCount":9,"composite":69.47,"weighted":65.91,"confidence":{"cross-verified":0,"single-source":3,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"MiniMax official"},"gpqa":{"level":"single-source","source":"MiniMax official"},"aime":{"level":"single-source","source":"MiniMax official"}}},{"slug":"step-3-5-flash","name":"Step-3.5-Flash","category":"llm","provider":"StepFun","type":"open","license":"Apache2","released":"2026.02","arch":"MoE · Reasoning(11B/196B)","priceIn":null,"priceOut":null,"maxIn":256,"maxOut":32,"tokPerSec":300,"ttft":null,"modelUrl":"https://huggingface.co/stepfun-ai/Step-3.5-Flash","scores":{"aime":97.3,"lcb":86.4,"sweV":74.4,"termBench":51},"benchCount":4,"composite":77.28,"weighted":48.87,"confidence":{"cross-verified":0,"single-source":2,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"StepFun official"},"gpqa":{"level":"single-source","source":"StepFun official"}}},{"slug":"gpt-5-4","name":"GPT-5.4","category":"llm","provider":"OpenAI","type":"closed","license":"Prop","released":"2026.03","arch":"Dense · Reasoning(Computer Use)","priceIn":2.5,"priceOut":15,"maxIn":1000,"maxOut":64,"tokPerSec":null,"ttft":null,"modelUrl":"https://openai.com/gpt-5","scores":{"gpqa":92.8,"hle":52.1,"arcAgi2":73.3,"swePro":57.7},"benchCount":4,"composite":68.97,"weighted":43.62,"confidence":{"cross-verified":1,"single-source":7,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"OpenAI official"},"gpqa":{"level":"single-source","source":"OpenAI official"},"aime":{"level":"single-source","source":"OpenAI official"},"hle":{"level":"single-source","source":"OpenAI official"},"swePro":{"level":"single-source","source":"OpenAI official"},"ifeval":{"level":"single-source","source":"OpenAI official"},"lcb":{"level":"single-source","source":"OpenAI official"},"arcAgi2":{"level":"cross-verified","source":"OpenAI + arcprize.org"}}},{"slug":"gpt-5-1","name":"GPT-5.1","category":"llm","provider":"OpenAI","type":"closed","license":"Prop","released":"2025.11","arch":"Dense · Reasoning(Adaptive)","priceIn":1.25,"priceOut":10,"maxIn":400,"maxOut":64,"tokPerSec":null,"ttft":1,"modelUrl":"https://openai.com/gpt-5","scores":{"gpqa":88.1,"aime":94,"hle":26,"arcAgi2":17,"sweV":74.9},"benchCount":5,"composite":60.0,"weighted":42.43,"confidence":{"cross-verified":0,"single-source":2,"self-reported":0},"confidenceDetail":{"gpqa":{"level":"single-source","source":"OpenAI official"},"aime":{"level":"single-source","source":"OpenAI official"}}},{"slug":"gemini-3-pro","name":"Gemini 3 Pro","category":"llm","provider":"Google","type":"closed","license":"Prop","released":"2025.11","arch":"Dense · Reasoning(DeepThink)","priceIn":2,"priceOut":12,"maxIn":1000,"maxOut":64,"tokPerSec":134,"ttft":3,"modelUrl":"https://deepmind.google/technologies/gemini/","scores":{"gpqa":91.9,"aime":95,"hle":37.5,"arcAgi2":31.1,"sweV":76.2,"mmmlu":85},"benchCount":6,"composite":69.45,"weighted":53.8,"confidence":{"cross-verified":0,"single-source":3,"self-reported":0},"confidenceDetail":{"gpqa":{"level":"single-source","source":"Google DeepMind"},"aime":{"level":"single-source","source":"Google DeepMind"},"hle":{"level":"single-source","source":"Google DeepMind"}}},{"slug":"claude-sonnet-4-5","name":"Claude Sonnet 4.5","category":"llm","provider":"Anthropic","type":"closed","license":"Prop","released":"2025.09","arch":"Dense · Reasoning(Hybrid)","priceIn":3,"priceOut":15,"maxIn":200,"maxOut":64,"tokPerSec":null,"ttft":1.5,"modelUrl":"https://anthropic.com/claude","scores":{"gpqa":83.4,"aime":100,"hle":30.8,"ifeval":90,"sweV":77.2,"termBench":51},"benchCount":6,"composite":72.07,"weighted":55.82,"confidence":{"cross-verified":2,"single-source":2,"self-reported":0},"confidenceDetail":{"gpqa":{"level":"cross-verified","source":"Anthropic + Vellum"},"aime":{"level":"cross-verified","source":"Anthropic + Vellum"},"swePro":{"level":"single-source","source":"Anthropic official"},"ifeval":{"level":"single-source","source":"Anthropic official"}}},{"slug":"gpt-5-nano","name":"GPT-5-Nano","category":"llm","provider":"OpenAI","type":"closed","license":"Prop","released":"2025.08","arch":"Dense · Non-Reasoning(Nano)","priceIn":0.05,"priceOut":0.4,"maxIn":400,"maxOut":16,"tokPerSec":null,"ttft":0.2,"modelUrl":"https://openai.com/gpt-5","scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":4,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"OpenAI official"},"gpqa":{"level":"single-source","source":"OpenAI official"},"aime":{"level":"single-source","source":"OpenAI official"},"ifeval":{"level":"single-source","source":"OpenAI official"}}},{"slug":"gemini-2-5-fl-lite","name":"Gemini 2.5 FL-Lite","category":"llm","provider":"Google","type":"closed","license":"Prop","released":"2025.06","arch":"Dense · Non-Reasoning+Thinking","priceIn":0.1,"priceOut":0.4,"maxIn":1000,"maxOut":64,"tokPerSec":null,"ttft":0.5,"modelUrl":"https://deepmind.google/technologies/gemini/flash/","scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":3,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"Google DeepMind"},"gpqa":{"level":"single-source","source":"Google DeepMind"},"aime":{"level":"single-source","source":"Google DeepMind"}}},{"slug":"qwen3-next-80b","name":"Qwen3-Next-80B","category":"llm","provider":"Alibaba","type":"open","license":"NCML","released":"2025.09","arch":"MoE+Hybrid · Reasoning(3B/80B)","priceIn":0.15,"priceOut":1.2,"maxIn":262,"maxOut":32,"tokPerSec":null,"ttft":null,"modelUrl":"https://huggingface.co/Qwen/Qwen3-Next-80B-A3B-Thinking","scores":{"mmluPro":82.7,"gpqa":77.2,"bfcl":49.7,"ifeval":88.9,"lcb":68.7,"mmmlu":81.3},"benchCount":6,"composite":74.75,"weighted":57.9,"confidence":{"cross-verified":0,"single-source":4,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"Alibaba Cloud official"},"gpqa":{"level":"single-source","source":"Alibaba Cloud official"},"aime":{"level":"single-source","source":"Alibaba Cloud official"},"ifeval":{"level":"single-source","source":"Alibaba Cloud official"}}},{"slug":"kimi-k3","name":"Kimi K3","category":"llm","provider":"Moonshot AI","type":"open","license":"Modified MIT","released":"2026.07","arch":"MoE · Sparse MoE 16/896 experts · Kimi Delta Attention(hybrid linear)+Attention Residuals · 2.8T total · MXFP4","priceIn":3,"priceOut":15,"maxIn":1000,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":"https://www.kimi.com/blog/kimi-k3","scores":{"gpqa":93.5,"aime":96.4,"hle":43.5,"lcb":68,"termBench":88.3},"benchCount":5,"composite":77.94,"weighted":55.11,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"gemini-3-flash-vlm","name":"Gemini 3 Flash","category":"vlm","provider":"Google","type":"closed","license":null,"released":null,"arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{"mmmu":87.6,"mmmuPro":80},"benchCount":2,"composite":83.8,"weighted":37.48,"confidence":{"cross-verified":0,"single-source":7,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"Google DeepMind model card"},"gpqa":{"level":"single-source","source":"Google DeepMind model card"},"aime":{"level":"single-source","source":"Google DeepMind model card"},"ifeval":{"level":"single-source","source":"Google DeepMind model card"},"lcb":{"level":"single-source","source":"Google DeepMind model card"},"sweV":{"level":"single-source","source":"Google DeepMind model card"},"mmmlu":{"level":"single-source","source":"Google DeepMind model card"}}},{"slug":"gemini-3-pro-vlm","name":"Gemini 3 Pro","category":"vlm","provider":"Google","type":"closed","license":null,"released":null,"arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{"mmmu":87.5,"mmmuPro":80},"benchCount":2,"composite":83.75,"weighted":37.45,"confidence":{"cross-verified":0,"single-source":3,"self-reported":0},"confidenceDetail":{"gpqa":{"level":"single-source","source":"Google DeepMind"},"aime":{"level":"single-source","source":"Google DeepMind"},"hle":{"level":"single-source","source":"Google DeepMind"}}},{"slug":"gemini-3-1-pro-vlm","name":"Gemini 3.1 Pro","category":"vlm","provider":"Google","type":"closed","license":null,"released":null,"arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{"mmmuPro":82},"benchCount":1,"composite":82.0,"weighted":25.93,"confidence":{"cross-verified":5,"single-source":5,"self-reported":0},"confidenceDetail":{"gpqa":{"level":"cross-verified","source":"Google DeepMind + PCMag + NxCode"},"aime":{"level":"cross-verified","source":"Google DeepMind + Vellum"},"hle":{"level":"cross-verified","source":"Google DeepMind + Vellum"},"arcAgi2":{"level":"cross-verified","source":"Google DeepMind + arcprize.org + NxCode + DataCamp"},"sweV":{"level":"cross-verified","source":"Google DeepMind + marc0.dev"},"mmluPro":{"level":"single-source","source":"Google DeepMind model card"},"swePro":{"level":"single-source","source":"Google DeepMind model card"},"ifeval":{"level":"single-source","source":"Google DeepMind model card"},"lcb":{"level":"single-source","source":"Google DeepMind model card"},"mmmlu":{"level":"single-source","source":"Google DeepMind model card"}}},{"slug":"gpt-5-2-vlm","name":"GPT-5.2","category":"vlm","provider":"OpenAI","type":"closed","license":null,"released":null,"arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{"mmmu":86.7},"benchCount":1,"composite":86.7,"weighted":27.42,"confidence":{"cross-verified":7,"single-source":2,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"cross-verified","source":"OpenAI + Vellum + DataCamp"},"gpqa":{"level":"cross-verified","source":"OpenAI + Vellum + DataCamp"},"aime":{"level":"cross-verified","source":"OpenAI + Vellum + DataCamp"},"ifeval":{"level":"cross-verified","source":"OpenAI + Vellum + DataCamp"},"lcb":{"level":"cross-verified","source":"OpenAI + Vellum + DataCamp"},"sweV":{"level":"cross-verified","source":"OpenAI + Vellum + DataCamp"},"hle":{"level":"single-source","source":"OpenAI official"},"arcAgi2":{"level":"cross-verified","source":"OpenAI + arcprize.org"},"mmmlu":{"level":"single-source","source":"OpenAI official"}}},{"slug":"gpt-5-original","name":"GPT-5 (original)","category":"vlm","provider":"OpenAI","type":"closed","license":null,"released":null,"arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{"mmmu":84.2},"benchCount":1,"composite":84.2,"weighted":26.63,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"claude-opus-4-6-vlm","name":"Claude Opus 4.6","category":"vlm","provider":"Anthropic","type":"closed","license":null,"released":null,"arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{"mmmuPro":85.1},"benchCount":1,"composite":85.1,"weighted":26.91,"confidence":{"cross-verified":6,"single-source":3,"self-reported":0},"confidenceDetail":{"gpqa":{"level":"cross-verified","source":"Anthropic + Vellum + DataCamp"},"aime":{"level":"cross-verified","source":"Anthropic + Vellum + NxCode"},"hle":{"level":"cross-verified","source":"Vellum + digitalapplied"},"arcAgi2":{"level":"cross-verified","source":"Vellum + llm-stats + NxCode + DataCamp"},"swePro":{"level":"cross-verified","source":"Anthropic + Scale AI SEAL"},"ifeval":{"level":"cross-verified","source":"Anthropic + Vellum"},"lcb":{"level":"single-source","source":"Anthropic official"},"sweV":{"level":"single-source","source":"Anthropic official"},"metacog":{"level":"single-source","source":"FINAL Bench dataset"}}},{"slug":"grok-4-heavy-vlm","name":"Grok 4 Heavy","category":"vlm","provider":"xAI","type":"closed","license":null,"released":null,"arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{"mmmu":76.5},"benchCount":1,"composite":76.5,"weighted":24.19,"confidence":{"cross-verified":0,"single-source":8,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"xAI official"},"gpqa":{"level":"single-source","source":"xAI official"},"aime":{"level":"single-source","source":"xAI official"},"hle":{"level":"single-source","source":"xAI official"},"arcAgi2":{"level":"single-source","source":"xAI official"},"swePro":{"level":"single-source","source":"xAI official"},"ifeval":{"level":"single-source","source":"xAI official"},"lcb":{"level":"single-source","source":"xAI official"}}},{"slug":"internvl3-5-241b","name":"InternVL3.5-241B","category":"vlm","provider":"OpenGVLab","type":"open","license":null,"released":null,"arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{"mmmu":77.7},"benchCount":1,"composite":77.7,"weighted":24.57,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"internvl3-78b","name":"InternVL3-78B","category":"vlm","provider":"OpenGVLab","type":"open","license":null,"released":null,"arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{"mmmu":72.2,"mathVista":79.6,"ai2d":89.7,"ocrBench":90.6,"hallusionBench":59.1,"mmBenchEN":89,"realWorldQA":78},"benchCount":7,"composite":79.74,"weighted":66.72,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"qwen2-5-vl-72b","name":"Qwen2.5-VL-72B","category":"vlm","provider":"Alibaba","type":"open","license":null,"released":null,"arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{"mmmu":70.2,"mathVista":74.8,"mmStar":70.8},"benchCount":3,"composite":71.93,"weighted":39.4,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"kimi-vl-a3b-thinking","name":"Kimi-VL-A3B-Thinking","category":"vlm","provider":"Moonshot","type":"open","license":null,"released":null,"arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{"mmmu":64,"mmmuPro":46.3,"mathVista":80.1,"mmStar":70.4,"mmBenchEN":84.4,"realWorldQA":70},"benchCount":6,"composite":69.2,"weighted":53.6,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"gpt-5-nano-vlm","name":"GPT-5-Nano","category":"vlm","provider":"OpenAI","type":null,"license":null,"released":null,"arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":4,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"OpenAI official"},"gpqa":{"level":"single-source","source":"OpenAI official"},"aime":{"level":"single-source","source":"OpenAI official"},"ifeval":{"level":"single-source","source":"OpenAI official"}}},{"slug":"gemini-2-5-fl-lite-vlm","name":"Gemini-2.5-FL-Lite","category":"vlm","provider":"Google","type":null,"license":null,"released":null,"arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"qwen3-vl-30b-a3b","name":"Qwen3-VL-30B-A3B","category":"vlm","provider":"Alibaba","type":null,"license":null,"released":null,"arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"qwen3-5-9b-vlm","name":"Qwen3.5-9B","category":"vlm","provider":"Alibaba","type":null,"license":null,"released":null,"arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":2,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"Alibaba Cloud official"},"gpqa":{"level":"single-source","source":"Alibaba Cloud official"}}},{"slug":"qwen3-5-4b-vlm","name":"Qwen3.5-4B","category":"vlm","provider":"Alibaba","type":null,"license":null,"released":null,"arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":2,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"Alibaba Cloud official"},"gpqa":{"level":"single-source","source":"Alibaba Cloud official"}}},{"slug":"gpt-5-4-agent","name":"GPT-5.4","category":"agent","provider":"OpenAI","type":null,"license":null,"released":null,"arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{"osWorld":75,"browseComp":82.7,"gdpvalAA":83},"benchCount":3,"composite":80.23,"weighted":43.95,"confidence":{"cross-verified":1,"single-source":7,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"OpenAI official"},"gpqa":{"level":"single-source","source":"OpenAI official"},"aime":{"level":"single-source","source":"OpenAI official"},"hle":{"level":"single-source","source":"OpenAI official"},"swePro":{"level":"single-source","source":"OpenAI official"},"ifeval":{"level":"single-source","source":"OpenAI official"},"lcb":{"level":"single-source","source":"OpenAI official"},"arcAgi2":{"level":"cross-verified","source":"OpenAI + arcprize.org"}}},{"slug":"claude-opus-4-6-agent","name":"Claude Opus 4.6","category":"agent","provider":"Anthropic","type":null,"license":null,"released":null,"arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{"osWorld":72.7,"tau2Bench":91.9,"browseComp":84,"termBench2":74.7,"gdpvalAA":1606,"swePro":45},"benchCount":6,"composite":329.05,"weighted":254.88,"confidence":{"cross-verified":6,"single-source":3,"self-reported":0},"confidenceDetail":{"gpqa":{"level":"cross-verified","source":"Anthropic + Vellum + DataCamp"},"aime":{"level":"cross-verified","source":"Anthropic + Vellum + NxCode"},"hle":{"level":"cross-verified","source":"Vellum + digitalapplied"},"arcAgi2":{"level":"cross-verified","source":"Vellum + llm-stats + NxCode + DataCamp"},"swePro":{"level":"cross-verified","source":"Anthropic + Scale AI SEAL"},"ifeval":{"level":"cross-verified","source":"Anthropic + Vellum"},"lcb":{"level":"single-source","source":"Anthropic official"},"sweV":{"level":"single-source","source":"Anthropic official"},"metacog":{"level":"single-source","source":"FINAL Bench dataset"}}},{"slug":"claude-sonnet-4-6-agent","name":"Claude Sonnet 4.6","category":"agent","provider":"Anthropic","type":null,"license":null,"released":null,"arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{"osWorld":72.5,"termBench2":53,"gdpvalAA":1633},"benchCount":3,"composite":586.17,"weighted":321.06,"confidence":{"cross-verified":3,"single-source":3,"self-reported":0},"confidenceDetail":{"gpqa":{"level":"cross-verified","source":"Anthropic + Vellum + NxCode"},"aime":{"level":"cross-verified","source":"Anthropic + Vellum + NxCode"},"arcAgi2":{"level":"cross-verified","source":"Anthropic + Vellum + NxCode"},"swePro":{"level":"single-source","source":"Anthropic official"},"ifeval":{"level":"single-source","source":"Anthropic official"},"sweV":{"level":"single-source","source":"Anthropic official"}}},{"slug":"gemini-3-1-pro-agent","name":"Gemini 3.1 Pro","category":"agent","provider":"Google","type":null,"license":null,"released":null,"arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{"tau2Bench":99.3,"browseComp":85.9,"termBench2":78.4,"gdpvalAA":1317},"benchCount":4,"composite":395.15,"weighted":249.91,"confidence":{"cross-verified":5,"single-source":5,"self-reported":0},"confidenceDetail":{"gpqa":{"level":"cross-verified","source":"Google DeepMind + PCMag + NxCode"},"aime":{"level":"cross-verified","source":"Google DeepMind + Vellum"},"hle":{"level":"cross-verified","source":"Google DeepMind + Vellum"},"arcAgi2":{"level":"cross-verified","source":"Google DeepMind + arcprize.org + NxCode + DataCamp"},"sweV":{"level":"cross-verified","source":"Google DeepMind + marc0.dev"},"mmluPro":{"level":"single-source","source":"Google DeepMind model card"},"swePro":{"level":"single-source","source":"Google DeepMind model card"},"ifeval":{"level":"single-source","source":"Google DeepMind model card"},"lcb":{"level":"single-source","source":"Google DeepMind model card"},"mmmlu":{"level":"single-source","source":"Google DeepMind model card"}}},{"slug":"gpt-5-2-agent","name":"GPT-5.2","category":"agent","provider":"OpenAI","type":null,"license":null,"released":null,"arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{"osWorld":38.2,"tau2Bench":82,"browseComp":77.9,"termBench2":64.9},"benchCount":4,"composite":65.75,"weighted":41.58,"confidence":{"cross-verified":7,"single-source":2,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"cross-verified","source":"OpenAI + Vellum + DataCamp"},"gpqa":{"level":"cross-verified","source":"OpenAI + Vellum + DataCamp"},"aime":{"level":"cross-verified","source":"OpenAI + Vellum + DataCamp"},"ifeval":{"level":"cross-verified","source":"OpenAI + Vellum + DataCamp"},"lcb":{"level":"cross-verified","source":"OpenAI + Vellum + DataCamp"},"sweV":{"level":"cross-verified","source":"OpenAI + Vellum + DataCamp"},"hle":{"level":"single-source","source":"OpenAI official"},"arcAgi2":{"level":"cross-verified","source":"OpenAI + arcprize.org"},"mmmlu":{"level":"single-source","source":"OpenAI official"}}},{"slug":"gpt-5-3-codex-agent","name":"GPT-5.3 Codex","category":"agent","provider":"OpenAI","type":null,"license":null,"released":null,"arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{"termBench2":77.3,"swePro":57},"benchCount":2,"composite":67.15,"weighted":30.03,"confidence":{"cross-verified":2,"single-source":0,"self-reported":0},"confidenceDetail":{"swePro":{"level":"cross-verified","source":"OpenAI + Scale AI SEAL"},"lcb":{"level":"cross-verified","source":"OpenAI + Scale AI SEAL"}}},{"slug":"gemini-3-flash-agent","name":"Gemini 3 Flash","category":"agent","provider":"Google","type":null,"license":null,"released":null,"arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{"termBench2":64.3},"benchCount":1,"composite":64.3,"weighted":20.33,"confidence":{"cross-verified":0,"single-source":7,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"Google DeepMind model card"},"gpqa":{"level":"single-source","source":"Google DeepMind model card"},"aime":{"level":"single-source","source":"Google DeepMind model card"},"ifeval":{"level":"single-source","source":"Google DeepMind model card"},"lcb":{"level":"single-source","source":"Google DeepMind model card"},"sweV":{"level":"single-source","source":"Google DeepMind model card"},"mmmlu":{"level":"single-source","source":"Google DeepMind model card"}}},{"slug":"qwen3-5-9b-agent","name":"Qwen3.5-9B","category":"agent","provider":"Alibaba","type":null,"license":null,"released":null,"arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{"tau2Bench":79.9,"screenSpotPro":66.1,"androidWorld":57.8},"benchCount":3,"composite":67.93,"weighted":37.21,"confidence":{"cross-verified":0,"single-source":2,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"Alibaba Cloud official"},"gpqa":{"level":"single-source","source":"Alibaba Cloud official"}}},{"slug":"qwen3-5-4b-agent","name":"Qwen3.5-4B","category":"agent","provider":"Alibaba","type":null,"license":null,"released":null,"arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{"tau2Bench":79.1,"screenSpotPro":50.3,"androidWorld":58.6},"benchCount":3,"composite":62.67,"weighted":34.32,"confidence":{"cross-verified":0,"single-source":2,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"Alibaba Cloud official"},"gpqa":{"level":"single-source","source":"Alibaba Cloud official"}}},{"slug":"minimax-m2-5-agent","name":"MiniMax-M2.5","category":"agent","provider":"MiniMax","type":null,"license":null,"released":null,"arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{"termBench2":42.2},"benchCount":1,"composite":42.2,"weighted":13.34,"confidence":{"cross-verified":0,"single-source":3,"self-reported":0},"confidenceDetail":{"mmluPro":{"level":"single-source","source":"MiniMax official"},"gpqa":{"level":"single-source","source":"MiniMax official"},"aime":{"level":"single-source","source":"MiniMax official"}}},{"slug":"gpt-image-1-5","name":"GPT Image 1.5","category":"image","provider":"OpenAI","type":null,"license":"Prop","released":"2026.01","arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"imagen-4","name":"Imagen 4","category":"image","provider":"Google","type":null,"license":"Prop","released":"2025.12","arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"flux-2-pro","name":"Flux 2 Pro","category":"image","provider":"BFL","type":null,"license":"Prop","released":"2026.01","arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"midjourney-v7","name":"Midjourney v7","category":"image","provider":"Midjourney","type":null,"license":"Prop","released":"2026.01","arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"flux-2-dev","name":"Flux 2 Dev","category":"image","provider":"BFL","type":null,"license":"Apache2","released":"2026.01","arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"ideogram-3-0","name":"Ideogram 3.0","category":"image","provider":"Ideogram","type":null,"license":"Prop","released":"2025.12","arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"dall-e-3-5","name":"DALL-E 3.5","category":"image","provider":"OpenAI","type":null,"license":"Prop","released":"2025.08","arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"nano-banana-2","name":"Nano Banana 2","category":"image","provider":"Google","type":null,"license":"Prop","released":"2025.12","arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"sd-3-5","name":"SD 3.5","category":"image","provider":"Stability AI","type":null,"license":"Open","released":"2024.10","arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"seedream-4-5","name":"Seedream 4.5","category":"image","provider":"ByteDance","type":null,"license":"Prop","released":"2025.11","arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"sora-2","name":"Sora 2","category":"video","provider":"OpenAI","type":null,"license":"Prop","released":"2025.09","arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"veo-3-1","name":"Veo 3.1","category":"video","provider":"Google","type":null,"license":"Prop","released":"2026.01","arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"runway-gen-4-5","name":"Runway Gen-4.5","category":"video","provider":"Runway","type":null,"license":"Prop","released":"2026.02","arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"kling-3-0","name":"Kling 3.0","category":"video","provider":"Kuaishou","type":null,"license":"Prop","released":"2026.02","arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"seedance-2-0","name":"Seedance 2.0","category":"video","provider":"ByteDance","type":null,"license":"Prop","released":"2026.01","arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"wan-2-6","name":"Wan 2.6","category":"video","provider":"Alibaba","type":null,"license":"Open","released":"2026.01","arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"pika-2-5","name":"Pika 2.5","category":"video","provider":"Pika","type":null,"license":"Prop","released":"2025.12","arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"luma-ray3","name":"Luma Ray3","category":"video","provider":"Luma","type":null,"license":"Prop","released":"2026.01","arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"ltx-2","name":"LTX-2","category":"video","provider":"Lightricks","type":null,"license":"Open","released":"2026.01","arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"hailuo-ai","name":"HaiLuo AI","category":"video","provider":"MiniMax","type":null,"license":"Prop","released":"2025.12","arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"suno-v4-5","name":"Suno v4.5","category":"music","provider":"Suno","type":null,"license":"Prop","released":"2025.11","arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"udio-v2","name":"Udio v2","category":"music","provider":"Udio","type":null,"license":"Prop","released":"2025.10","arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"gemini-music","name":"Gemini Music","category":"music","provider":"Google","type":null,"license":"Prop","released":"2026.01","arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"musicgen-large","name":"MusicGen Large","category":"music","provider":"Meta","type":null,"license":"Open","released":"2024.06","arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"stable-audio-2-0","name":"Stable Audio 2.0","category":"music","provider":"Stability AI","type":null,"license":"Open","released":"2024.04","arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"jasco","name":"JASCO","category":"music","provider":"Meta","type":null,"license":"Open","released":"2024.12","arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"riffusion-v2","name":"Riffusion v2","category":"music","provider":"Riffusion","type":null,"license":"Prop","released":"2025.06","arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null},{"slug":"loudme","name":"Loudme","category":"music","provider":"Loudme","type":null,"license":"Prop","released":"2025.09","arch":null,"priceIn":null,"priceOut":null,"maxIn":null,"maxOut":null,"tokPerSec":null,"ttft":null,"modelUrl":null,"scores":{},"benchCount":0,"composite":null,"weighted":null,"confidence":{"cross-verified":0,"single-source":0,"self-reported":0},"confidenceDetail":null}]}}