Leaderboards
Benchmarks
Every score carries its origin.
Arena Creative Writing168Arena Instruction Following168Arena Text (overall)168IFEval16LiveBench Instruction Following58LiveBench Language58
LiveBench Language
% average over LiveBench language comprehension tasksProvenance58 modelsscores Jun 25, 2026
1Claude Fable 5AnthropicScore90.68Score / $2ScoredJun 25, 20262Claude Fable 5.1AnthropicScore89.5Score / $2ScoredJun 25, 20263GPT-6 AstraOpenAIScore89.43Score / $4ScoredJun 25, 20264Claude Opus 5AnthropicScore88.69Score / $4ScoredJun 25, 20265Gemini 3.8 FlashGoogleScore87.79Score / $47ScoredJun 25, 20266GPT-5.6 SolOpenAIScore87.68Score / $18ScoredJun 25, 20267GPT-5.5OpenAIScore87.36Score / $6ScoredJun 25, 20268Claude Opus 5.5AnthropicScore86.27Score / $4ScoredJun 25, 20269Union AlphastealthScore85.89ScoredJun 25, 202610Kimi K3Moonshot AI2.8TScore85.53Score / $9Score / GB VRAM~0.049ScoredJun 25, 202611Gemini 3.7 FlashGoogleScore85.46Score / $46ScoredJun 25, 202612GPT-6 SolOpenAIScore85.3Score / $17ScoredJun 25, 202613Gemini 3.5 FlashGoogleScore84.58Score / $19ScoredJun 25, 202614Gemini 3.6 FlashGoogleScore83.9Score / $45ScoredJun 25, 202615Grok 4.6xAIScore83.7Score / $14ScoredJun 25, 202616Claude Sonnet 5.5AnthropicScore83.43Score / $8ScoredJun 25, 202617Claude Opus 4.6AnthropicScore83.27Score / $3ScoredJun 25, 202618GPT-5.6 TerraOpenAIScore82.89Score / $14ScoredJun 25, 202619Grok 4.5xAIScore82.8Score / $14ScoredJun 25, 202620Muse Spark 1.3MetaScore82.79Score / $19ScoredJun 25, 202621GPT-5.4OpenAIScore82.63Score / $11ScoredJun 25, 202622Claude Opus 4.5AnthropicScore81.26Score / $3ScoredJun 25, 202623DeepSeek V4.1 FlashDeepSeek763BScore81.19Score / $203Score / GB VRAM~0.17ScoredJun 25, 202624DeepSeek V4 Flash Vision ExpDeepSeek305BScore80.36Score / $124Score / GB VRAM~0.42ScoredJun 25, 202625Grok 4.7xAIScore80.14Score / $13ScoredJun 25, 202626GLM 5.3Z.aiScore79.86Score / $70ScoredJun 25, 202627GPT-5.2 ChatOpenAIScore79.81Score / $11ScoredJun 25, 202628Qwen3.7 MaxQwenScore79.74Score / $21ScoredJun 25, 202629Qwen3.8 MaxQwenScore79.69Score / $16ScoredJun 25, 202630Qwen3 MaxQwenScore79.69Score / $20ScoredJun 25, 202631Claude Opus 4.8AnthropicScore79.66Score / $3ScoredJun 25, 202632Muse Spark 1.2MetaScore78.57Score / $18ScoredJun 25, 202633DeepSeek V4 ProDeepSeek1.6TScore78.13Score / $50Score / GB VRAM~0.078ScoredJun 25, 202634Kimi K2.7 CodeMoonshot AI1.1TScore77.91Score / $26Score / GB VRAM~0.12ScoredJun 25, 202635Claude Opus 4.7AnthropicScore77.91Score / $3ScoredJun 25, 202636GLM 5.3 FlashZ.ai321BScore77.31Score / $309Score / GB VRAM~0.38ScoredJun 25, 202637MiniMax M3MiniMax427BScore76.84Score / $80Score / GB VRAM~0.29ScoredJun 25, 202638GLM 5.2Z.ai753BScore76.24Score / $84Score / GB VRAM~0.16ScoredJun 25, 202639Claude Sonnet 4.6AnthropicScore76.1Score / $5ScoredJun 25, 202640Kimi K2.6Moonshot AI1.1TScore75.14Score / $41Score / GB VRAM~0.11ScoredJun 25, 202641Qwen3.6 PlusQwenScore74.99Score / $38ScoredJun 25, 202642Claude Sonnet 5AnthropicScore74.97Score / $7ScoredJun 25, 202643Qwen3.8 FlashQwenScore74.64Score / $195ScoredJun 25, 202644Qwen3.8 27BQwen27.8BScore74.35Score / $50Score / GB VRAM~4.2ScoredJun 25, 202645Muse Spark 1.1MetaScore74.34Score / $17ScoredJun 25, 202646GPT-6 LunaOpenAIScore73.83Score / $295ScoredJun 25, 202647GPT-5.2-CodexOpenAIScore73.68Score / $5ScoredJun 25, 202648Grok 4.3xAIScore73.58Score / $29ScoredJun 25, 202649InklingThinking Machines952BScore73.46Score / $18Score / GB VRAM~0.12ScoredJun 25, 202650GPT-5.6 LunaOpenAIScore72.57Score / $121ScoredJun 25, 202651Grok Build 0.1xAIScore72.46Score / $36ScoredJun 25, 202652Gemini 3.5 Flash LiteGoogleScore71.82Score / $57ScoredJun 25, 202653GPT-5.4 MiniOpenAIScore70.95Score / $32ScoredJun 25, 202654Nemotron 3 UltraNVIDIA561BScore70.81Score / $32Score / GB VRAM~0.2ScoredJun 25, 202655DeepSeek V4 FlashDeepSeek291BScore70.12Score / $838Score / GB VRAM~0.38ScoredJun 25, 202656Ox AlphastealthScore66.14ScoredJun 25, 202657Qwen3.6 27BQwen27.8BScore63.3Score / $32Score / GB VRAM~3.6ScoredJun 25, 202658GPT-5.4 NanoOpenAIScore62.51Score / $100ScoredJun 25, 2026