Leaderboards
Benchmarks
Every score carries its origin.
Arena Creative Writing168Arena Instruction Following168Arena Text (overall)168IFEval16LiveBench Instruction Following58LiveBench Language58
Arena Creative Writing
Arena Score on creative-writing prompts — human preference, not a prose-craft rubricProvenance100 of 168 modelsscores Aug 10, 2026 – Sep 25, 2026
1Claude Opus 5.5AnthropicScore1521Score / $76ScoredSep 25, 20262Claude Fable 5AnthropicScore1504Score / $30ScoredSep 13, 20263Gemini 3.7 FlashGoogleScore1492Score / $796ScoredSep 25, 20264Claude Fable 5.1AnthropicScore1484Score / $30ScoredSep 25, 20265Claude Opus 4.7AnthropicScore1484Score / $59ScoredSep 25, 20266Gemini 3.8 FlashGoogleScore1482Score / $790ScoredSep 25, 20267Gemini 3.1 Pro Preview Custom ToolsGoogleScore1480Score / $247ScoredSep 25, 20268Claude Opus 4.6AnthropicScore1478Score / $59ScoredSep 25, 20269Gemini 3.6 FlashGoogleScore1473Score / $786ScoredAug 10, 202610Claude Opus 5AnthropicScore1472Score / $59ScoredSep 25, 202611GPT-5.6 SolOpenAIScore1470Score / $294ScoredSep 25, 202612Qwen3.8 MaxQwenScore1467Score / $296ScoredSep 25, 202613Gemini 3.5 FlashGoogleScore1466Score / $326ScoredSep 25, 202614Muse Spark 1.1MetaScore1465Score / $345ScoredSep 25, 202615Grok 4.20xAIScore1464Score / $586ScoredSep 25, 202616Claude Opus 4.8AnthropicScore1462Score / $58ScoredSep 25, 202617Claude Opus 4.5AnthropicScore1461Score / $58ScoredSep 25, 202618Kimi K3Moonshot AI2.8TScore1458Score / $149Score / GB VRAM~0.83ScoredSep 25, 202619GLM 5.3Z.aiScore1456Score / $1278ScoredSep 25, 202620GPT-5.5OpenAIScore1456Score / $97ScoredSep 25, 202621GLM 5.2Z.ai753BScore1455Score / $1599Score / GB VRAM~3.1ScoredSep 25, 202622Claude Sonnet 4.5AnthropicScore1454Score / $97ScoredSep 25, 202623GPT-6 AstraOpenAIScore1453Score / $58ScoredSep 25, 202624Grok 4.5xAIScore1452Score / $242ScoredSep 25, 202625GPT-6 SolOpenAIScore1451Score / $290ScoredSep 25, 202626Muse Spark 1.3MetaScore1450Score / $341ScoredSep 25, 202627Claude Sonnet 4.6AnthropicScore1449Score / $97ScoredSep 25, 202628Muse Spark 1.2MetaScore1448Score / $341ScoredSep 25, 202629GLM 5.1Z.ai754BScore1448Score / $478Score / GB VRAM~3ScoredSep 25, 202630GLM 5Z.ai754BScore1447Score / $754Score / GB VRAM~3ScoredSep 25, 202631Grok 4.20 Multi-AgentxAIScore1447Score / $579ScoredSep 25, 202632DeepSeek V4 ProDeepSeek1.6TScore1446Score / $923Score / GB VRAM~1.4ScoredSep 25, 202633Grok 4.6xAIScore1445Score / $241ScoredSep 25, 202634Gemini 2.5 ProGoogleScore1443Score / $289ScoredSep 25, 202635Claude Opus 4.1AnthropicScore1442Score / $19ScoredSep 25, 202636Qwen3.7 MaxQwenScore1441Score / $384ScoredSep 25, 202637DeepSeek V4.1 FlashDeepSeek763BScore1440Score / $3600Score / GB VRAM~3ScoredSep 25, 202638Qwen3.6 Max PreviewQwenScore1436Score / $233ScoredSep 25, 202639Gemini 3.5 Flash LiteGoogleScore1435Score / $1148ScoredSep 25, 202640MiMo-V2.5-ProXiaomi1TScore1435Score / $2356Score / GB VRAM~2.2ScoredSep 25, 202641GPT-5.4OpenAIScore1434Score / $191ScoredSep 25, 202642GPT-5.2 ChatOpenAIScore1433Score / $205ScoredSep 25, 202643GLM 5.3 FlashZ.ai321BScore1433Score / $5733Score / GB VRAM~7.1ScoredSep 25, 202644Kimi K2.6Moonshot AI1.1TScore1433Score / $784Score / GB VRAM~2.1ScoredSep 25, 202645Qwen3.7 PlusQwenScore1432Score / $1118ScoredSep 25, 202646GPT-5.6 TerraOpenAIScore1428Score / $238ScoredSep 25, 202647Grok 4.7xAIScore1427Score / $238ScoredSep 25, 202648Kimi K2 ThinkingMoonshot AI1.1TScore1423Score / $569Score / GB VRAM~2.1ScoredSep 13, 202649Grok 4.3xAIScore1423Score / $569ScoredSep 25, 202650Gemma 4 31BGoogle31.3BScore1419Score / $4301Score / GB VRAM~72ScoredSep 25, 202651Hy3Tencent299BScore1419Score / $2688Score / GB VRAM~7.5ScoredSep 25, 202652Claude Opus 4AnthropicScore1417Score / $19ScoredSep 25, 202653Gemini 3.1 Flash Lite PreviewGoogleScore1413Score / $1885ScoredSep 25, 202654GPT-5.6 LunaOpenAIScore1412Score / $2353ScoredSep 25, 202655DeepSeek V3.2 ExpDeepSeek685BScore1409Score / $3435Score / GB VRAM~3.3ScoredSep 25, 202656GPT-5.3 ChatOpenAIScore1408Score / $101ScoredSep 25, 202657DeepSeek V4 FlashDeepSeek291BScore1408Score / $16815Score / GB VRAM~7.7ScoredSep 25, 202658GPT-5.1OpenAIScore1408Score / $282ScoredSep 25, 202659Qwen3.6 PlusQwenScore1407Score / $722ScoredSep 25, 202660MiniMax M3MiniMax427BScore1406Score / $1465Score / GB VRAM~5.2ScoredSep 25, 202661Qwen3.5 397B A17BQwen403BScore1405Score / $601Score / GB VRAM~5.5ScoredSep 25, 202662DeepSeek V3.1 TerminusDeepSeek685BScore1405Score / $1405Score / GB VRAM~3.3ScoredSep 25, 202663GLM 4.7Z.ai358BScore1404Score / $802Score / GB VRAM~6.2ScoredSep 25, 202664GLM 5V TurboZ.aiScore1403Score / $351ScoredSep 25, 202665GPT-6 LunaOpenAIScore1403Score / $5611ScoredSep 25, 202666GPT-4.1OpenAIScore1402Score / $175ScoredSep 25, 202667GPT-5.4 MiniOpenAIScore1402Score / $623ScoredSep 25, 202668DeepSeek V3.2DeepSeek685BScore1400Score / $4523Score / GB VRAM~3.2ScoredSep 25, 202669GLM 4.6Z.ai357BScore1400Score / $800Score / GB VRAM~6.2ScoredSep 25, 202670Gemma 4 26B A4BGoogle25.8BScore1399Score / $6360Score / GB VRAM~86ScoredSep 25, 202671Mistral Medium 3.5Mistral AI128BScore1398Score / $186Score / GB VRAM~17ScoredSep 25, 202672MiMo-V2.6-FlashXiaomiScore1394Score / $5242ScoredSep 25, 202673Gemini 2.5 FlashGoogleScore1394Score / $1115ScoredSep 25, 202674MiMo-V2.5Xiaomi311BScore1394Score / $5856Score / GB VRAM~7.1ScoredSep 25, 202675Qwen3 MaxQwenScore1393Score / $357ScoredSep 25, 202676DeepSeek V3 0324DeepSeek685BScore1390Score / $1544Score / GB VRAM~3.2ScoredSep 25, 202677Kimi K2.5Moonshot AI1.1TScore1390Score / $618Score / GB VRAM~2.1ScoredSep 25, 202678GPT-5OpenAIScore1389Score / $139ScoredSep 25, 202679R1 0528DeepSeek685BScore1389Score / $646Score / GB VRAM~3.2ScoredSep 25, 202680Claude Sonnet 4AnthropicScore1388Score / $93ScoredSep 25, 202681DeepSeek V3.1DeepSeek685BScore1388Score / $1461Score / GB VRAM~3.2ScoredSep 25, 202682Claude Haiku 4.5AnthropicScore1387Score / $277ScoredSep 25, 202683InklingThinking Machines952BScore1384Score / $342Score / GB VRAM~2.3ScoredSep 25, 202684o3OpenAIScore1383Score / $173ScoredSep 25, 202685o1OpenAIScore1381Score / $23ScoredSep 25, 202686Qwen3 235B A22B Instruct 2507Qwen235BScore1378Score / $3937Score / GB VRAM~9.3ScoredSep 25, 202687Mistral Large 3 2512Mistral AIScore1373Score / $915ScoredSep 25, 202688GLM 4.5Z.ai358BScore1372Score / $624Score / GB VRAM~6.1ScoredSep 25, 202689Claude Sonnet 5AnthropicScore1370Score / $137ScoredSep 13, 202690Qwen3 235B A22B Thinking 2507Qwen235BScore1370Score / $596Score / GB VRAM~9.2ScoredSep 25, 202691Qwen3.5-122B-A10BQwen125BScore1366Score / $657Score / GB VRAM~17ScoredSep 25, 202692Muse Glimmer 30BMeta29.8BScore1365Score / $1241Score / GB VRAM~73ScoredSep 25, 202693Qwen3.8 27BQwen27.8BScore1364Score / $916Score / GB VRAM~78ScoredSep 25, 202694MiniMax M2.7MiniMax229BScore1363Score / $1623Score / GB VRAM~9.5ScoredSep 25, 202695Qwen3 Coder 480B A35BQwen480BScore1362Score / $1362Score / GB VRAM~4.5ScoredSep 25, 202696Qwen3.5-27BQwen27.8BScore1358Score / $870Score / GB VRAM~77ScoredSep 25, 202697Qwen3 VL 235B A22B InstructQwen236BScore1357Score / $1542Score / GB VRAM~9.1ScoredSep 25, 202698MiniMax M2.5MiniMax229BScore1357Score / $1428Score / GB VRAM~9.4ScoredSep 25, 202699Hy3 previewTencent299BScore1354Score / $2257Score / GB VRAM~7.2ScoredSep 25, 2026100GLM 4.6VZ.ai108BScore1350Score / $1500Score / GB VRAM~20ScoredSep 25, 2026
Showing 1–100 of 168 models on this board.