Leaderboards

Benchmarks

Every score carries its origin.

Reasoning374Coding337Math221Agentic375Chat636Speech778
Arena Creative Writing168Arena Instruction Following168Arena Text (overall)168IFEval16LiveBench Instruction Following58LiveBench Language58

Arena Text (overall)

Arena Score (Bradley-Terry, style-controlled) from human pairwise votes — not a percentage
Provenance100 of 168 modelsscores Aug 10, 2026 – Sep 25, 2026
1Claude Opus 5.5AnthropicScore1509Score / $75ScoredSep 25, 20262Claude Fable 5AnthropicScore1506Score / $30ScoredSep 13, 20263Claude Fable 5.1AnthropicScore1501Score / $30ScoredSep 25, 20264Claude Opus 4.6AnthropicScore1498Score / $60ScoredSep 25, 20265Muse Spark 1.2MetaScore1496Score / $352ScoredSep 25, 20266Claude Opus 4.7AnthropicScore1495Score / $60ScoredSep 25, 20267Muse Spark 1.3MetaScore1494Score / $352ScoredSep 25, 20268Gemini 3.8 FlashGoogleScore1492Score / $796ScoredSep 25, 20269Claude Opus 5AnthropicScore1491Score / $60ScoredSep 25, 202610Muse Spark 1.1MetaScore1491Score / $351ScoredSep 25, 202611Gemini 3.7 FlashGoogleScore1488Score / $794ScoredSep 25, 202612Kimi K3Moonshot AI2.8TScore1488Score / $152Score / GB VRAM~0.85ScoredSep 25, 202613Gemini 3.1 Pro Preview Custom ToolsGoogleScore1487Score / $248ScoredSep 25, 202614Gemini 3.6 FlashGoogleScore1484Score / $791ScoredAug 10, 202615GPT-5.6 SolOpenAIScore1483Score / $297ScoredSep 25, 202616GLM 5.3Z.aiScore1480Score / $1298ScoredSep 25, 202617Qwen3.8 MaxQwenScore1479Score / $299ScoredSep 25, 202618GPT-6 AstraOpenAIScore1478Score / $59ScoredSep 25, 202619Gemini 3.5 FlashGoogleScore1477Score / $328ScoredSep 25, 202620GPT-5.5OpenAIScore1477Score / $98ScoredSep 25, 202621DeepSeek V4.1 FlashDeepSeek763BScore1477Score / $3691Score / GB VRAM~3.1ScoredSep 25, 202622GPT-5.2 ChatOpenAIScore1476Score / $211ScoredSep 25, 202623GLM 5.2Z.ai753BScore1476Score / $1621Score / GB VRAM~3.1ScoredSep 25, 202624Qwen3.7 MaxQwenScore1475Score / $393ScoredSep 25, 202625Grok 4.20xAIScore1475Score / $590ScoredSep 25, 202626GLM 5.3 FlashZ.ai321BScore1474Score / $5897Score / GB VRAM~7.3ScoredSep 25, 202627Claude Opus 4.8AnthropicScore1474Score / $59ScoredSep 25, 202628Claude Sonnet 4.6AnthropicScore1472Score / $98ScoredSep 25, 202629Grok 4.20 Multi-AgentxAIScore1470Score / $588ScoredSep 25, 202630Claude Opus 4.5AnthropicScore1470Score / $59ScoredSep 25, 202631MiMo-V2.5-ProXiaomi1TScore1467Score / $2410Score / GB VRAM~2.3ScoredSep 25, 202632Grok 4.5xAIScore1465Score / $244ScoredSep 25, 202633GLM 5.1Z.ai754BScore1465Score / $483Score / GB VRAM~3.1ScoredSep 25, 202634GPT-5.6 TerraOpenAIScore1465Score / $244ScoredSep 25, 202635GPT-5.4OpenAIScore1465Score / $195ScoredSep 25, 202636DeepSeek V4 ProDeepSeek1.6TScore1464Score / $935Score / GB VRAM~1.5ScoredSep 25, 202637Kimi K2.6Moonshot AI1.1TScore1461Score / $799Score / GB VRAM~2.2ScoredSep 25, 202638Qwen3.6 Max PreviewQwenScore1460Score / $237ScoredSep 25, 202639GPT-6 SolOpenAIScore1457Score / $291ScoredSep 25, 202640GLM 5Z.ai754BScore1457Score / $759Score / GB VRAM~3.1ScoredSep 25, 202641Hy3Tencent299BScore1457Score / $2759Score / GB VRAM~7.7ScoredSep 25, 202642Gemini 3.5 Flash LiteGoogleScore1456Score / $1165ScoredSep 25, 202643Qwen3.7 PlusQwenScore1456Score / $1137ScoredSep 25, 202644Claude Sonnet 4.5AnthropicScore1455Score / $97ScoredSep 25, 202645GPT-5.6 LunaOpenAIScore1454Score / $2423ScoredSep 25, 202646MiMo-V2.6-FlashXiaomiScore1454Score / $5465ScoredSep 25, 202647Grok 4.6xAIScore1453Score / $242ScoredSep 25, 202648Gemma 4 31BGoogle31.3BScore1453Score / $4402Score / GB VRAM~74ScoredSep 25, 202649Kimi K2 ThinkingMoonshot AI1.1TScore1450Score / $580Score / GB VRAM~2.2ScoredSep 13, 202650GPT-5.3 ChatOpenAIScore1450Score / $104ScoredSep 25, 202651Claude Opus 4.1AnthropicScore1448Score / $19ScoredSep 25, 202652GPT-5.4 MiniOpenAIScore1447Score / $643ScoredSep 25, 202653Gemini 2.5 ProGoogleScore1446Score / $289ScoredSep 25, 202654Qwen3.6 PlusQwenScore1443Score / $740ScoredSep 25, 202655GPT-6 LunaOpenAIScore1442Score / $5770ScoredSep 25, 202656InklingThinking Machines952BScore1442Score / $356Score / GB VRAM~2.4ScoredSep 25, 202657Qwen3.5 397B A17BQwen403BScore1442Score / $616Score / GB VRAM~5.7ScoredSep 25, 202658Grok 4.3xAIScore1442Score / $577ScoredSep 25, 202659GLM 4.7Z.ai358BScore1441Score / $824Score / GB VRAM~6.4ScoredSep 25, 202660MiniMax M3MiniMax427BScore1440Score / $1500Score / GB VRAM~5.4ScoredSep 25, 202661Grok 4.7xAIScore1439Score / $240ScoredSep 25, 202662GPT-5.1OpenAIScore1438Score / $288ScoredSep 25, 202663Qwen3.8 27BQwen27.8BScore1438Score / $965Score / GB VRAM~82ScoredSep 25, 202664Gemma 4 26B A4BGoogle25.8BScore1438Score / $6535Score / GB VRAM~88ScoredSep 25, 202665DeepSeek V4 FlashDeepSeek291BScore1436Score / $17153Score / GB VRAM~7.8ScoredSep 25, 202666Qwen3 MaxQwenScore1435Score / $368ScoredSep 25, 202667MiMo-V2.5Xiaomi311BScore1434Score / $6023Score / GB VRAM~7.3ScoredSep 25, 202668GLM 5V TurboZ.aiScore1433Score / $358ScoredSep 25, 202669Gemini 3.1 Flash Lite PreviewGoogleScore1433Score / $1910ScoredSep 25, 202670o3OpenAIScore1432Score / $179ScoredSep 25, 202671Kimi K2.5Moonshot AI1.1TScore1430Score / $636Score / GB VRAM~2.1ScoredSep 25, 202672GPT-5OpenAIScore1427Score / $143ScoredSep 25, 202673Mistral Medium 3.5Mistral AI128BScore1426Score / $190Score / GB VRAM~18ScoredSep 25, 202674DeepSeek V3.2DeepSeek685BScore1425Score / $4602Score / GB VRAM~3.3ScoredSep 25, 202675Muse Glimmer 30BMeta29.8BScore1424Score / $1295Score / GB VRAM~76ScoredSep 25, 202676GLM 4.6Z.ai357BScore1424Score / $814Score / GB VRAM~6.3ScoredSep 25, 202677Qwen3 235B A22B Instruct 2507Qwen235BScore1422Score / $4064Score / GB VRAM~9.6ScoredSep 25, 202678DeepSeek V3.2 ExpDeepSeek685BScore1422Score / $3468Score / GB VRAM~3.3ScoredSep 25, 202679R1 0528DeepSeek685BScore1422Score / $661Score / GB VRAM~3.3ScoredSep 25, 202680DeepSeek V3.1DeepSeek685BScore1417Score / $1492Score / GB VRAM~3.3ScoredSep 25, 202681Qwen3.5-122B-A10BQwen125BScore1416Score / $681Score / GB VRAM~18ScoredSep 25, 202682GPT-4.1OpenAIScore1415Score / $177ScoredSep 25, 202683MiniMax M2.7MiniMax229BScore1415Score / $1684Score / GB VRAM~9.8ScoredSep 25, 202684DeepSeek V3.1 TerminusDeepSeek685BScore1415Score / $1415Score / GB VRAM~3.3ScoredSep 25, 202685Claude Opus 4AnthropicScore1414Score / $19ScoredSep 25, 202686Claude Haiku 4.5AnthropicScore1414Score / $283ScoredSep 25, 202687Qwen3 VL 235B A22B InstructQwen236BScore1413Score / $1606Score / GB VRAM~9.5ScoredSep 25, 202688Mistral Large 3 2512Mistral AIScore1413Score / $942ScoredSep 25, 202689GLM 4.5Z.ai358BScore1411Score / $642Score / GB VRAM~6.2ScoredSep 25, 202690Gemini 2.5 FlashGoogleScore1409Score / $1128ScoredSep 25, 202691Hy3 previewTencent299BScore1409Score / $2349Score / GB VRAM~7.5ScoredSep 25, 202692Qwen3.5-27BQwen27.8BScore1409Score / $903Score / GB VRAM~80ScoredSep 25, 202693Inkling SmallThinking Machines266BScore1405Score / $1171Score / GB VRAM~8.4ScoredSep 25, 202694o1OpenAIScore1402Score / $23ScoredSep 25, 202695GPT-5.4 NanoOpenAIScore1401Score / $2242ScoredSep 25, 202696Qwen3 235B A22B Thinking 2507Qwen235BScore1400Score / $609Score / GB VRAM~9.4ScoredSep 25, 202697Qwen3 Next 80B A3B InstructQwen81.3BScore1399Score / $1794Score / GB VRAM~27ScoredSep 25, 202698Qwen3.5-FlashQwenScore1396Score / $5370ScoredSep 25, 202699DeepSeek V3 0324DeepSeek685BScore1396Score / $1551Score / GB VRAM~3.2ScoredSep 25, 2026100Qwen3 VL 235B A22B ThinkingQwen236BScore1395Score / $353Score / GB VRAM~9.4ScoredSep 25, 2026

Showing 1–100 of 168 models on this board.