Leaderboards
Benchmarks
Every score carries its origin.
Arena Coding
Arena Score on coding prompts within the text arena — not a percentageProvenance100 of 168 modelsscores Aug 10, 2026 – Sep 25, 2026
1Claude Fable 5AnthropicScore1552Score / $31ScoredSep 13, 20262Claude Opus 4.6AnthropicScore1548Score / $62ScoredSep 25, 20263Claude Opus 5.5AnthropicScore1547Score / $77ScoredSep 25, 20264Claude Opus 4.7AnthropicScore1547Score / $62ScoredSep 25, 20265GPT-6 AstraOpenAIScore1542Score / $62ScoredSep 25, 20266Kimi K3Moonshot AI2.8TScore1541Score / $157Score / GB VRAM~0.88ScoredSep 25, 20267Muse Spark 1.3MetaScore1539Score / $362ScoredSep 25, 20268Muse Spark 1.2MetaScore1535Score / $361ScoredSep 25, 20269Claude Opus 5AnthropicScore1533Score / $61ScoredSep 25, 202610Muse Spark 1.1MetaScore1532Score / $361ScoredSep 25, 202611DeepSeek V4.1 FlashDeepSeek763BScore1532Score / $3830Score / GB VRAM~3.2ScoredSep 25, 202612Gemini 3.8 FlashGoogleScore1532Score / $817ScoredSep 25, 202613GPT-5.6 SolOpenAIScore1531Score / $306ScoredSep 25, 202614Claude Opus 4.8AnthropicScore1530Score / $61ScoredSep 25, 202615Claude Fable 5.1AnthropicScore1529Score / $31ScoredSep 25, 202616Claude Sonnet 4.6AnthropicScore1529Score / $102ScoredSep 25, 202617MiMo-V2.6-FlashXiaomiScore1525Score / $5733ScoredSep 25, 202618Qwen3.7 MaxQwenScore1525Score / $407ScoredSep 25, 202619GPT-6 SolOpenAIScore1524Score / $305ScoredSep 25, 202620Claude Opus 4.5AnthropicScore1523Score / $61ScoredSep 25, 202621Gemini 3.6 FlashGoogleScore1523Score / $812ScoredAug 10, 202622GLM 5.3 FlashZ.ai321BScore1523Score / $6090Score / GB VRAM~7.5ScoredSep 25, 202623GLM 5.3Z.aiScore1522Score / $1335ScoredSep 25, 202624MiMo-V2.5-ProXiaomi1TScore1522Score / $2498Score / GB VRAM~2.4ScoredSep 25, 202625Gemini 3.1 Pro Preview Custom ToolsGoogleScore1521Score / $253ScoredSep 25, 202626Qwen3.8 MaxQwenScore1520Score / $307ScoredSep 25, 202627Gemini 3.7 FlashGoogleScore1519Score / $810ScoredSep 25, 202628GPT-5.6 TerraOpenAIScore1516Score / $253ScoredSep 25, 202629Kimi K2.6Moonshot AI1.1TScore1516Score / $829Score / GB VRAM~2.3ScoredSep 25, 202630GPT-5.2 ChatOpenAIScore1515Score / $216ScoredSep 25, 202631GPT-6 LunaOpenAIScore1515Score / $6058ScoredSep 25, 202632GPT-5.5OpenAIScore1513Score / $101ScoredSep 25, 202633Grok 4.5xAIScore1513Score / $252ScoredSep 25, 202634GLM 5.1Z.ai754BScore1513Score / $499Score / GB VRAM~3.2ScoredSep 25, 202635Qwen3.6 Max PreviewQwenScore1512Score / $245ScoredSep 25, 202636Claude Sonnet 4.5AnthropicScore1512Score / $101ScoredSep 25, 202637GLM 5.2Z.ai753BScore1512Score / $1662Score / GB VRAM~3.2ScoredSep 25, 202638GPT-5.4OpenAIScore1511Score / $201ScoredSep 25, 202639Grok 4.20 Multi-AgentxAIScore1508Score / $603ScoredSep 25, 202640Grok 4.20xAIScore1508Score / $603ScoredSep 25, 202641Grok 4.6xAIScore1507Score / $251ScoredSep 25, 202642Gemini 3.5 FlashGoogleScore1507Score / $335ScoredSep 25, 202643DeepSeek V4 ProDeepSeek1.6TScore1506Score / $962Score / GB VRAM~1.5ScoredSep 25, 202644Claude Opus 4.1AnthropicScore1505Score / $20ScoredSep 25, 202645Qwen3.8 27BQwen27.8BScore1504Score / $1009Score / GB VRAM~86ScoredSep 25, 202646Gemini 3.5 Flash LiteGoogleScore1503Score / $1203ScoredSep 25, 202647Kimi K2.5Moonshot AI1.1TScore1503Score / $668Score / GB VRAM~2.3ScoredSep 25, 202648Kimi K2 ThinkingMoonshot AI1.1TScore1502Score / $601Score / GB VRAM~2.3ScoredSep 13, 202649Gemma 4 31BGoogle31.3BScore1502Score / $4551Score / GB VRAM~76ScoredSep 25, 202650Hy3Tencent299BScore1501Score / $2843Score / GB VRAM~8ScoredSep 25, 202651GPT-5.6 LunaOpenAIScore1501Score / $2501ScoredSep 25, 202652Qwen3.7 PlusQwenScore1500Score / $1172ScoredSep 25, 202653GPT-5.3 ChatOpenAIScore1498Score / $107ScoredSep 25, 202654GLM 5Z.ai754BScore1498Score / $780Score / GB VRAM~3.2ScoredSep 25, 202655Qwen3.6 PlusQwenScore1496Score / $767ScoredSep 25, 202656MiniMax M3MiniMax427BScore1495Score / $1557Score / GB VRAM~5.6ScoredSep 25, 202657GPT-5.4 MiniOpenAIScore1495Score / $664ScoredSep 25, 202658Qwen3.5 397B A17BQwen403BScore1491Score / $637Score / GB VRAM~5.9ScoredSep 25, 202659InklingThinking Machines952BScore1491Score / $368Score / GB VRAM~2.5ScoredSep 25, 202660MiMo-V2.5Xiaomi311BScore1491Score / $6263Score / GB VRAM~7.6ScoredSep 25, 202661Grok 4.7xAIScore1488Score / $248ScoredSep 25, 202662GLM 5V TurboZ.aiScore1488Score / $372ScoredSep 25, 202663Grok 4.3xAIScore1488Score / $595ScoredSep 25, 202664DeepSeek V4 FlashDeepSeek291BScore1484Score / $17727Score / GB VRAM~8.1ScoredSep 25, 202665GLM 4.7Z.ai358BScore1484Score / $848Score / GB VRAM~6.6ScoredSep 25, 202666Gemma 4 26B A4BGoogle25.8BScore1483Score / $6739Score / GB VRAM~91ScoredSep 25, 202667Claude Haiku 4.5AnthropicScore1482Score / $296ScoredSep 25, 202668Qwen3 MaxQwenScore1480Score / $380ScoredSep 25, 202669Muse Glimmer 30BMeta29.8BScore1480Score / $1345Score / GB VRAM~79ScoredSep 25, 202670MiniMax M2.7MiniMax229BScore1480Score / $1761Score / GB VRAM~10ScoredSep 25, 202671Mistral Medium 3.5Mistral AI128BScore1478Score / $197Score / GB VRAM~18ScoredSep 25, 202672GPT-5.1OpenAIScore1476Score / $295ScoredSep 25, 202673Inkling SmallThinking Machines266BScore1473Score / $1227Score / GB VRAM~8.8ScoredSep 25, 202674Qwen3 235B A22B Instruct 2507Qwen235BScore1472Score / $4205Score / GB VRAM~9.9ScoredSep 25, 202675DeepSeek V3.2DeepSeek685BScore1469Score / $4746Score / GB VRAM~3.4ScoredSep 25, 202676Mistral Large 3 2512Mistral AIScore1469Score / $979ScoredSep 25, 202677DeepSeek V3.2 ExpDeepSeek685BScore1467Score / $3577Score / GB VRAM~3.4ScoredSep 25, 202678Claude Opus 4AnthropicScore1465Score / $20ScoredSep 25, 202679Gemini 2.5 ProGoogleScore1464Score / $293ScoredSep 25, 202680R1 0528DeepSeek685BScore1464Score / $681Score / GB VRAM~3.4ScoredSep 25, 202681Qwen3 VL 235B A22B InstructQwen236BScore1463Score / $1662Score / GB VRAM~9.8ScoredSep 25, 202682GPT-5OpenAIScore1462Score / $146ScoredSep 25, 202683o3OpenAIScore1460Score / $183ScoredSep 25, 202684GPT-5.4 NanoOpenAIScore1460Score / $2336ScoredSep 25, 202685Qwen3.5-122B-A10BQwen125BScore1459Score / $701Score / GB VRAM~18ScoredSep 25, 202686GLM 4.6Z.ai357BScore1458Score / $833Score / GB VRAM~6.5ScoredSep 25, 202687Gemini 3.1 Flash Lite PreviewGoogleScore1458Score / $1944ScoredSep 25, 202688Qwen3 Coder 480B A35BQwen480BScore1457Score / $1457Score / GB VRAM~4.8ScoredSep 25, 202689GPT-4.1OpenAIScore1456Score / $182ScoredSep 25, 202690GLM 4.5Z.ai358BScore1455Score / $661Score / GB VRAM~6.4ScoredSep 25, 202691Hy3 previewTencent299BScore1453Score / $2422Score / GB VRAM~7.7ScoredSep 25, 202692Qwen3 VL 235B A22B ThinkingQwen236BScore1452Score / $367Score / GB VRAM~9.8ScoredSep 25, 202693Qwen3.5-27BQwen27.8BScore1451Score / $930Score / GB VRAM~83ScoredSep 25, 202694Step 3.5 FlashStepFun199BScore1449Score / $4831Score / GB VRAM~12ScoredSep 25, 202695Claude Sonnet 4AnthropicScore1449Score / $97ScoredSep 25, 202696DeepSeek V3.1DeepSeek685BScore1447Score / $1524Score / GB VRAM~3.4ScoredSep 25, 202697MiniMax M2.5MiniMax229BScore1446Score / $1523Score / GB VRAM~10ScoredSep 25, 202698Qwen3 Next 80B A3B InstructQwen81.3BScore1444Score / $1852Score / GB VRAM~28ScoredSep 25, 202699Qwen3 235B A22B Thinking 2507Qwen235BScore1442Score / $627Score / GB VRAM~9.7ScoredSep 25, 2026100DeepSeek V3.1 TerminusDeepSeek685BScore1440Score / $1440Score / GB VRAM~3.3ScoredSep 25, 2026
Showing 1–100 of 168 models on this board.