[{"modelId":"o3","rating":1151.3779287263492,"rank":1,"num_battles":1076,"rating_q025":1127.1508457746786,"rating_q975":1174.3957832775193,"variance":null,"cost_per_100_calls_usd":3.5438},{"modelId":"Claude-4.1-Opus","rating":1126.3532524689376,"rank":2,"num_battles":341,"rating_q025":1098.9556194029617,"rating_q975":1154.837547542123,"variance":null,"cost_per_100_calls_usd":28.648},{"modelId":"GPT-5","rating":1122.5188843645722,"rank":3,"num_battles":357,"rating_q025":1085.8245154231768,"rating_q975":1157.0663541330218,"variance":null,"cost_per_100_calls_usd":2.9752},{"modelId":"Gemini-3-Pro-Preview","rating":1086.3325465626408,"rank":4,"num_battles":215,"rating_q025":1051.8638049003746,"rating_q975":1133.4082199103375,"variance":null,"cost_per_100_calls_usd":3.3879},{"modelId":"GPT-5.1","rating":1079.672379931729,"rank":5,"num_battles":200,"rating_q025":1029.6520277253167,"rating_q975":1127.3105741444924,"variance":null,"cost_per_100_calls_usd":2.5648},{"modelId":"Claude-4-Opus","rating":1071.7486973245118,"rank":6,"num_battles":1403,"rating_q025":1054.4818424482155,"rating_q975":1094.0012620766615,"variance":null,"cost_per_100_calls_usd":28.4471},{"modelId":"GPT-5-mini","rating":1067.1098895478674,"rank":7,"num_battles":342,"rating_q025":1037.6837256766291,"rating_q975":1101.362714499022,"variance":null,"cost_per_100_calls_usd":0.589},{"modelId":"Gemini-2.5-Pro","rating":1061.267593810923,"rank":8,"num_battles":1253,"rating_q025":1038.6396950038572,"rating_q975":1081.6137102799942,"variance":null,"cost_per_100_calls_usd":2.8685},{"modelId":"Grok-4","rating":1044.571125247441,"rank":9,"num_battles":465,"rating_q025":1019.3626144304872,"rating_q975":1077.6443525674927,"variance":null,"cost_per_100_calls_usd":5.732},{"modelId":"Deepseek-R1-0528","rating":1041.8412987548516,"rank":10,"num_battles":1340,"rating_q025":1026.379572763093,"rating_q975":1060.8209094761676,"variance":null,"cost_per_100_calls_usd":0.7853},{"modelId":"GPT-OSS-120B","rating":1036.5017542645915,"rank":11,"num_battles":350,"rating_q025":1008.7709271198796,"rating_q975":1068.2050206196457,"variance":null,"cost_per_100_calls_usd":0.0989},{"modelId":"Qwen3-235B-A22B-Thinking-2507","rating":1036.2789141482078,"rank":12,"num_battles":381,"rating_q025":1003.3372633439888,"rating_q975":1069.455355360025,"variance":null,"cost_per_100_calls_usd":0.2377},{"modelId":"o4-mini","rating":1028.5553925129395,"rank":13,"num_battles":1712,"rating_q025":1013.097642999501,"rating_q975":1046.484079931444,"variance":null,"cost_per_100_calls_usd":2.4773},{"modelId":"Claude-4-Sonnet","rating":1025.205465465613,"rank":14,"num_battles":1336,"rating_q025":1004.8131444384621,"rating_q975":1039.7337115301236,"variance":null,"cost_per_100_calls_usd":5.6999},{"modelId":"Qwen3-235B-A22B-2507","rating":1020.5209790938372,"rank":15,"num_battles":263,"rating_q025":982.1061963178382,"rating_q975":1071.8902165312293,"variance":null,"cost_per_100_calls_usd":0.2129},{"modelId":"GPT-4.1","rating":1018.5843886213381,"rank":16,"num_battles":2011,"rating_q025":1004.4438231836951,"rating_q975":1034.0110928808028,"variance":null,"cost_per_100_calls_usd":2.7261},{"modelId":"GPT-4.1-mini","rating":1014.585805498339,"rank":17,"num_battles":1416,"rating_q025":996.1516528407979,"rating_q975":1031.7681566274223,"variance":null,"cost_per_100_calls_usd":0.5522},{"modelId":"Qwen3-30B-A3B-Instruct-2507","rating":1010.1989495896634,"rank":18,"num_battles":375,"rating_q025":971.3246976966576,"rating_q975":1039.9495270864834,"variance":null,"cost_per_100_calls_usd":0.2784},{"modelId":"Gemini-2.5-Pro-Preview","rating":1009.6882693653926,"rank":19,"num_battles":1044,"rating_q025":990.2447593449193,"rating_q975":1032.4624425176853,"variance":null,"cost_per_100_calls_usd":2.9302},{"modelId":"GLM-4.5","rating":1006.9434167887279,"rank":20,"num_battles":371,"rating_q025":972.8886563441282,"rating_q975":1040.778710804728,"variance":null,"cost_per_100_calls_usd":0.9986},{"modelId":"Deepseek-R1","rating":1006.8506439057406,"rank":21,"num_battles":1814,"rating_q025":990.5091136747874,"rating_q975":1024.2239310594446,"variance":null,"cost_per_100_calls_usd":0.744},{"modelId":"Deepseek-V3","rating":1003.4881441523811,"rank":22,"num_battles":1989,"rating_q025":991.475275129337,"rating_q975":1020.2389141630598,"variance":null,"cost_per_100_calls_usd":0.3745},{"modelId":"Qwen3-235B-A22B","rating":1002.5899428580932,"rank":23,"num_battles":1672,"rating_q025":987.8008684866934,"rating_q975":1021.9461964004042,"variance":null,"cost_per_100_calls_usd":0.3603},{"modelId":"Kimi-K2","rating":1001.8875949017187,"rank":24,"num_battles":501,"rating_q025":978.523696092663,"rating_q975":1029.8166383329649,"variance":null,"cost_per_100_calls_usd":0.7471},{"modelId":"Grok-3","rating":989.3482237409862,"rank":25,"num_battles":1935,"rating_q025":975.0764031389684,"rating_q975":1007.0457132408825,"variance":null,"cost_per_100_calls_usd":4.1394},{"modelId":"QwQ-32B","rating":977.5366315062888,"rank":26,"num_battles":1775,"rating_q025":955.9378802582492,"rating_q975":993.2753436086057,"variance":null,"cost_per_100_calls_usd":0.1093},{"modelId":"Claude-3-7-Sonnet","rating":966.4162650074346,"rank":27,"num_battles":1961,"rating_q025":948.9101534894645,"rating_q975":981.154466171073,"variance":null,"cost_per_100_calls_usd":5.7391},{"modelId":"Gemini-2.5-Flash","rating":965.9602361107354,"rank":28,"num_battles":1472,"rating_q025":950.4808663142911,"rating_q975":981.4827422816211,"variance":null,"cost_per_100_calls_usd":0.7081},{"modelId":"Olmo-3.1-32B-Instruct","rating":963.5932586246543,"rank":29,"num_battles":128,"rating_q025":914.1216427410296,"rating_q975":1009.1314289102336,"variance":null,"cost_per_100_calls_usd":0.17},{"modelId":"Qwen3-32B","rating":962.7216968345899,"rank":30,"num_battles":1662,"rating_q025":945.7305783957307,"rating_q975":978.8691423622967,"variance":null,"cost_per_100_calls_usd":0.1655},{"modelId":"Gemini-2.5-Flash-Preview","rating":932.7643886440749,"rank":31,"num_battles":1370,"rating_q025":913.4688307952363,"rating_q975":951.7143214570232,"variance":null,"cost_per_100_calls_usd":0.7247},{"modelId":"GPT-OSS-20B","rating":927.5760470644791,"rank":32,"num_battles":370,"rating_q025":894.9049207314397,"rating_q975":959.2633305617214,"variance":null,"cost_per_100_calls_usd":0.0546},{"modelId":"GPT-5-nano","rating":902.5921491006892,"rank":33,"num_battles":330,"rating_q025":860.0412747074714,"rating_q975":933.5732355070469,"variance":null,"cost_per_100_calls_usd":0.1167},{"modelId":"Mistral-Small-3.1","rating":889.7638801379572,"rank":34,"num_battles":1663,"rating_q025":875.8660889282506,"rating_q975":908.5008836935349,"variance":null,"cost_per_100_calls_usd":0.0647},{"modelId":"Mistral-Medium-3","rating":884.7929589152496,"rank":35,"num_battles":1761,"rating_q025":871.6269978669983,"rating_q975":902.9317086420431,"variance":null,"cost_per_100_calls_usd":0.6545},{"modelId":"Minimax-M1","rating":879.9097531390266,"rank":36,"num_battles":726,"rating_q025":856.7982036421033,"rating_q975":904.096747568523,"variance":null,"cost_per_100_calls_usd":0.566},{"modelId":"Llama-4-Maverick","rating":844.3286227130445,"rank":37,"num_battles":1749,"rating_q025":832.1670223763682,"rating_q975":862.5548808002302,"variance":null,"cost_per_100_calls_usd":0.2016},{"modelId":"Llama-4-Scout","rating":829.7737302958208,"rank":38,"num_battles":2112,"rating_q025":815.0842412312433,"rating_q975":844.754349995399,"variance":null,"cost_per_100_calls_usd":0.1082}]