{"config":"agent_task_outcome_explicit","id":"arena-agent-task-outcome-explicit","title":"Agents · explicit task outcomes","category":"Coding and agents","metric":"IPS score","direction":"higher","precision":6,"publishedAt":"2026-11-04","interval":"95% confidence interval","observationLabel":"Observations","description":"Agent Arena signal scores from the official dataset. Each signal is a separate measurement; it is not a success percentage.","limitations":["The population is the models in this source board, including variants and models without an available API price.","Ranks use the published confidence bounds: one plus the number of models whose lower bound exceeds this model’s upper bound. Shared ranks are not proof of equal performance."],"omissions":[],"rows":[{"name":"Claude Opus 5.5 (Max)","organization":"anthropic","score":0.21149356176594047,"low":0.179357962001337,"high":0.24362916153054395,"observations":4352,"sourceRank":1,"publishedAt":"2026-11-04","rank":1},{"name":"Claude Sonnet 5.5 (Max)","organization":"anthropic","score":0.20209566377851307,"low":0.15804490586482778,"high":0.24614642169219836,"observations":2205,"sourceRank":2,"publishedAt":"2026-11-04","rank":1},{"name":"Gemini 4 Carbon (High)","organization":"google","score":0.18021330451009543,"low":0.16765017841636015,"high":0.1927764306038307,"observations":22690,"sourceRank":3,"publishedAt":"2026-11-04","rank":1},{"name":"GPT-6 Astra (Max)","organization":"openai","score":0.14168478239180493,"low":0.10810667832717881,"high":0.17526288645643107,"observations":6555,"sourceRank":4,"publishedAt":"2026-11-04","rank":2},{"name":"Gemini 4 Argon (High)","organization":"google","score":0.13367054239686382,"low":0.11832783954970404,"high":0.1490132452440236,"observations":26963,"sourceRank":5,"publishedAt":"2026-11-04","rank":4},{"name":"Claude Fable 5.1 (Max)","organization":"anthropic","score":0.13290452897514804,"low":0.09731146871581937,"high":0.16849758923447672,"observations":7487,"sourceRank":6,"publishedAt":"2026-11-04","rank":2},{"name":"GPT-6.1 Sol (Max)","organization":"openai","score":0.12189536605953533,"low":0.10232450206780276,"high":0.1414662300512679,"observations":12235,"sourceRank":7,"publishedAt":"2026-11-04","rank":4},{"name":"Muse Wonder Full (Max)","organization":"meta","score":0.11441618342358784,"low":0.09645504590297763,"high":0.13237732094419805,"observations":16254,"sourceRank":8,"publishedAt":"2026-11-04","rank":4},{"name":"Claude Opus 5 (Max)","organization":"anthropic","score":0.11383359574643004,"low":0.09007710770617515,"high":0.1375900837866849,"observations":18539,"sourceRank":9,"publishedAt":"2026-11-04","rank":4},{"name":"GPT-6.1 Terra (Max)","organization":"openai","score":0.11133568121676851,"low":0.08396425748383043,"high":0.1387071049497066,"observations":10041,"sourceRank":10,"publishedAt":"2026-11-04","rank":4},{"name":"Hy4","organization":"tencent","score":0.07395107067915485,"low":0.04171790028299062,"high":0.10618424107531908,"observations":4692,"sourceRank":11,"publishedAt":"2026-11-04","rank":6},{"name":"Muse Wonder (Max)","organization":"meta","score":0.063306868138293,"low":0.034035150620647936,"high":0.09257858565593807,"observations":8073,"sourceRank":12,"publishedAt":"2026-11-04","rank":9},{"name":"Claude Fable 5 (High)","organization":"anthropic","score":0.056482940642869295,"low":0.03453185095986057,"high":0.07843403032587802,"observations":36809,"sourceRank":13,"publishedAt":"2026-11-04","rank":11},{"name":"MiMo-V2.6-Flash","organization":"xiaomi","score":0.055038666979800754,"low":0.03247617064465151,"high":0.07760116331495,"observations":13651,"sourceRank":14,"publishedAt":"2026-11-04","rank":11},{"name":"Grok 4.7 (xHigh)","organization":"xai","score":0.05340563607082549,"low":0.022369648071358737,"high":0.08444162407029224,"observations":9073,"sourceRank":15,"publishedAt":"2026-11-04","rank":10},{"name":"MiMo-V2.6-Pro","organization":"xiaomi","score":0.05249096182145585,"low":0.033001529012479616,"high":0.07198039463043208,"observations":21044,"sourceRank":16,"publishedAt":"2026-11-04","rank":11},{"name":"Hy4 preview","organization":"tencent","score":0.045471430230065835,"low":0.02988861420040745,"high":0.06105424625972422,"observations":36269,"sourceRank":17,"publishedAt":"2026-11-04","rank":11},{"name":"Qwen3.8-Max","organization":"alibaba","score":0.04337652472589215,"low":0.029049078825861954,"high":0.05770397062592235,"observations":42576,"sourceRank":18,"publishedAt":"2026-11-04","rank":11},{"name":"Kimi K3.1 Pro (Max)","organization":"moonshot","score":0.040177311583900144,"low":0.011919049866976221,"high":0.06843557330082406,"observations":10202,"sourceRank":19,"publishedAt":"2026-11-04","rank":11},{"name":"Muse Spark 1.3 (Max)","organization":"meta","score":0.03978172592860707,"low":0.02194853969570843,"high":0.05761491216150571,"observations":44614,"sourceRank":20,"publishedAt":"2026-11-04","rank":11},{"name":"Kimi K3 (Max)","organization":"moonshot","score":0.03745844957598532,"low":0.024495719224395373,"high":0.05042117992757526,"observations":99584,"sourceRank":21,"publishedAt":"2026-11-04","rank":11},{"name":"GPT-6 Sol (Max)","organization":"openai","score":0.03537579593082363,"low":0.0011952955220454475,"high":0.06955629633960181,"observations":6126,"sourceRank":22,"publishedAt":"2026-11-04","rank":11},{"name":"DeepSeek-V4.1-Flash-20261017 (Max)","organization":"deepseek","score":0.033770635542903604,"low":0.020558983648460163,"high":0.046982287437347045,"observations":35025,"sourceRank":23,"publishedAt":"2026-11-04","rank":11},{"name":"GLM-5.3 (Max)","organization":"zai","score":0.031921715064640144,"low":0.014407295046224167,"high":0.049436135083056124,"observations":58030,"sourceRank":24,"publishedAt":"2026-11-04","rank":11},{"name":"GLM-5.2 (Max)","organization":"zai","score":0.025384904092306537,"low":0.008490384577870436,"high":0.04227942360674264,"observations":57302,"sourceRank":25,"publishedAt":"2026-11-04","rank":11},{"name":"GLM-5.3-Flash","organization":"zai","score":0.02316026123458137,"low":0.007056477190424359,"high":0.03926404527873838,"observations":59023,"sourceRank":26,"publishedAt":"2026-11-04","rank":12},{"name":"GPT-5.6 Sol (xHigh)","organization":"openai","score":0.01972641080506421,"low":-0.0027761600471522055,"high":0.042228981657280625,"observations":31493,"sourceRank":27,"publishedAt":"2026-11-04","rank":11},{"name":"Claude Haiku 5.5 (High)","organization":"anthropic","score":0.014337576670139351,"low":-0.010339297006869853,"high":0.03901445034714855,"observations":16264,"sourceRank":28,"publishedAt":"2026-11-04","rank":12},{"name":"Gemini 3.8 Flash (High)","organization":"google","score":0.011984819854856,"low":-0.005741232900954397,"high":0.029710872610666395,"observations":28414,"sourceRank":29,"publishedAt":"2026-11-04","rank":17},{"name":"MiniMax-M3.1-Flash","organization":"minimax","score":0.010977437656641498,"low":-0.02691631902940167,"high":0.048871194342684665,"observations":4626,"sourceRank":30,"publishedAt":"2026-11-04","rank":11},{"name":"Qwen3.8-Flash-Next","organization":"alibaba","score":0.0038617632620192577,"low":-0.013583786587476231,"high":0.021307313111514747,"observations":41035,"sourceRank":31,"publishedAt":"2026-11-04","rank":21},{"name":"Claude Sonnet 5 (High)","organization":"anthropic","score":-0.00023031566170395035,"low":-0.026370346707314626,"high":0.025909715383906726,"observations":27267,"sourceRank":32,"publishedAt":"2026-11-04","rank":18},{"name":"GPT-6 Luna (Max)","organization":"openai","score":-0.0016733232130206055,"low":-0.01166455135952699,"high":0.00831790493348578,"observations":110526,"sourceRank":33,"publishedAt":"2026-11-04","rank":25},{"name":"Step 5 Preview","organization":"stepfun","score":-0.004597378664433771,"low":-0.0367406858718601,"high":0.027545928542992558,"observations":7033,"sourceRank":34,"publishedAt":"2026-11-04","rank":18},{"name":"Grok 4.6 (xHigh)","organization":"xai","score":-0.018889394962588124,"low":-0.0452347226059293,"high":0.007455932680753046,"observations":17841,"sourceRank":35,"publishedAt":"2026-11-04","rank":25},{"name":"Qwen3.8-27B","organization":"alibaba","score":-0.024189875172226594,"low":-0.04373416758673576,"high":-0.004645582757717429,"observations":31813,"sourceRank":36,"publishedAt":"2026-11-04","rank":28},{"name":"DeepSeek-V4-Pro (High, 20260813)","organization":"deepseek","score":-0.027969022249890285,"low":-0.06619103385597766,"high":0.010252989356197097,"observations":3402,"sourceRank":37,"publishedAt":"2026-11-04","rank":24},{"name":"Gemini 3.7 Flash (High)","organization":"google","score":-0.031183989063973683,"low":-0.05058645203636648,"high":-0.011781526091580886,"observations":43575,"sourceRank":38,"publishedAt":"2026-11-04","rank":31},{"name":"Mistral Large 4","organization":"mistral","score":-0.043109638111884285,"low":-0.0896888387098177,"high":0.003469562486049127,"observations":2389,"sourceRank":39,"publishedAt":"2026-11-04","rank":26},{"name":"GPT-6 Luna (Medium)","organization":"openai","score":-0.06233421395172256,"low":-0.07537690122498866,"high":-0.04929152667845646,"observations":36762,"sourceRank":40,"publishedAt":"2026-11-04","rank":37},{"name":"Muse Spark 1.1","organization":"meta","score":-0.06533497389393284,"low":-0.08079411891820587,"high":-0.0498758288696598,"observations":86947,"sourceRank":41,"publishedAt":"2026-11-04","rank":37},{"name":"GPT-5.6 Terra (xHigh)","organization":"openai","score":-0.07317894845219897,"low":-0.09122249769638814,"high":-0.05513539920800979,"observations":29566,"sourceRank":42,"publishedAt":"2026-11-04","rank":38},{"name":"Spark-X2.6","organization":"synlan","score":-0.07879191231815952,"low":-0.11642158200895666,"high":-0.041162242627362365,"observations":4124,"sourceRank":43,"publishedAt":"2026-11-04","rank":35},{"name":"Muse Spark 1.2 (xHigh)","organization":"meta","score":-0.09177292039144591,"low":-0.11319321873022363,"high":-0.07035262205266819,"observations":30983,"sourceRank":44,"publishedAt":"2026-11-04","rank":39},{"name":"Qwen3.7-Plus","organization":"alibaba","score":-0.09209794675657634,"low":-0.12842570065369824,"high":-0.05577019285945446,"observations":15743,"sourceRank":45,"publishedAt":"2026-11-04","rank":38},{"name":"Qwen3.7-Max","organization":"alibaba","score":-0.1033311012443866,"low":-0.13063560040388483,"high":-0.07602660208488837,"observations":28968,"sourceRank":46,"publishedAt":"2026-11-04","rank":40},{"name":"Beam 1.0 501B A23B","organization":"reflection","score":-0.10630847598414583,"low":-0.1649004329060944,"high":-0.047716519062197246,"observations":5050,"sourceRank":47,"publishedAt":"2026-11-04","rank":37},{"name":"MiMo-V2.5-Pro","organization":"xiaomi","score":-0.10848581413865699,"low":-0.1323008374324847,"high":-0.08467079084482927,"observations":29180,"sourceRank":48,"publishedAt":"2026-11-04","rank":41},{"name":"Hy3","organization":"tencent","score":-0.11669518244157151,"low":-0.14774881719909233,"high":-0.08564154768405069,"observations":13853,"sourceRank":49,"publishedAt":"2026-11-04","rank":41},{"name":"Gemini 3.1 Pro Preview","organization":"google","score":-0.1191676809515333,"low":-0.14738835468019149,"high":-0.0909470072228751,"observations":69499,"sourceRank":50,"publishedAt":"2026-11-04","rank":42},{"name":"GPT-6 Luna (Low)","organization":"openai","score":-0.15258397283787795,"low":-0.17091229895242407,"high":-0.13425564672333182,"observations":22963,"sourceRank":51,"publishedAt":"2026-11-04","rank":48},{"name":"Mistral Medium 3.5","organization":"mistral","score":-0.17048247729413146,"low":-0.21041335172611297,"high":-0.13055160286214995,"observations":7353,"sourceRank":52,"publishedAt":"2026-11-04","rank":46},{"name":"MiniMax-M3","organization":"minimax","score":-0.1767009130002879,"low":-0.20539040627709954,"high":-0.14801141972347626,"observations":26992,"sourceRank":53,"publishedAt":"2026-11-04","rank":50},{"name":"NVIDIA-Nemotron-3.5-Super-VL-120B-A12B (NVFP4)","organization":"nvidia","score":-0.18614678643992866,"low":-0.20925646739374942,"high":-0.1630371054861079,"observations":10821,"sourceRank":54,"publishedAt":"2026-11-04","rank":50},{"name":"Nemotron 3 Ultra","organization":"nvidia","score":-0.23366183648938088,"low":-0.3106112541416808,"high":-0.15671241883708098,"observations":4275,"sourceRank":55,"publishedAt":"2026-11-04","rank":50},{"name":"Inkling-Small","organization":"thinky","score":-0.2642321487563861,"low":-0.3130201084501613,"high":-0.21544418906261092,"observations":7078,"sourceRank":56,"publishedAt":"2026-11-04","rank":55},{"name":"Inkling","organization":"thinky","score":-0.27038406829348605,"low":-0.3093890978484847,"high":-0.2313790387384874,"observations":24084,"sourceRank":57,"publishedAt":"2026-11-04","rank":55},{"name":"Solar Pro 4","organization":"upstage","score":-0.3078141788602917,"low":-0.3661767214920177,"high":-0.24945163622856575,"observations":4588,"sourceRank":58,"publishedAt":"2026-11-04","rank":55}],"source":{"id":"arena","publisher":"Arena (lmarena-ai)","href":"https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset","licence":"CC-BY-4.0","licenceUrl":"https://creativecommons.org/licenses/by/4.0/","revision":"8b6deb69ace01ee8dcda9e46eba7a5e8a808ec53","fetchedAt":"2026-11-05T06:35:05.520Z","evidence":[{"file":"a7a159f557a1e594a5d41218fe8d53c63c63daa39b655e51d6fc5b482a17d059.json","sha256":"a7a159f557a1e594a5d41218fe8d53c63c63daa39b655e51d6fc5b482a17d059","url":"https://huggingface.co/api/datasets/lmarena-ai/leaderboard-dataset","revision":null,"role":"metadata"},{"file":"88a46d472f35274fb6b7a852c3d16c8ee8007d51b37f290161537bec26895ee4.txt","sha256":"88a46d472f35274fb6b7a852c3d16c8ee8007d51b37f290161537bec26895ee4","url":"https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/raw/8b6deb69ace01ee8dcda9e46eba7a5e8a808ec53/README.md","revision":"8b6deb69ace01ee8dcda9e46eba7a5e8a808ec53","role":"licence"},{"file":"5e92ac469b5228e8dd69af6e80b180527dc30728d814c5a8ec7cf17f7b8e1bd1.parquet","sha256":"5e92ac469b5228e8dd69af6e80b180527dc30728d814c5a8ec7cf17f7b8e1bd1","url":"https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/resolve/8b6deb69ace01ee8dcda9e46eba7a5e8a808ec53/agent_task_outcome_explicit/latest-00000-of-00001.parquet","revision":"8b6deb69ace01ee8dcda9e46eba7a5e8a808ec53","role":"rows"}]},"status":"fresh"}