Leaderboards
Benchmarks
Every score carries its origin.
Arena Creative Writing168Arena Instruction Following168Arena Text (overall)168IFEval16LiveBench Instruction Following58LiveBench Language58
LiveBench Instruction Following
% average over LiveBench constrained-rewriting tasksProvenance58 modelsscores Jun 25, 2026
1Gemini 3.8 FlashGoogleScore81.41Score / $43ScoredJun 25, 20262Gemini 3.7 FlashGoogleScore79.93Score / $43ScoredJun 25, 20263Muse Spark 1.3MetaScore78Score / $18ScoredJun 25, 20264Qwen3.8 FlashQwenScore77.11Score / $202ScoredJun 25, 20265Claude Fable 5AnthropicScore75.77Score / $2ScoredJun 25, 20266Gemini 3.5 FlashGoogleScore75.6Score / $17ScoredJun 25, 20267GPT-6 AstraOpenAIScore75.58Score / $3ScoredJun 25, 20268Gemini 3.6 FlashGoogleScore75.37Score / $40ScoredJun 25, 20269Grok 4.7xAIScore75.28Score / $13ScoredJun 25, 202610Muse Spark 1.2MetaScore74.33Score / $17ScoredJun 25, 202611Qwen3 MaxQwenScore74.08Score / $19ScoredJun 25, 202612Qwen3.8 MaxQwenScore74.08Score / $15ScoredJun 25, 202613Qwen3.7 MaxQwenScore74.04Score / $20ScoredJun 25, 202614Nemotron 3 UltraNVIDIA561BScore73.45Score / $33Score / GB VRAM~0.21ScoredJun 25, 202615Claude Fable 5.1AnthropicScore72.99Score / $1ScoredJun 25, 202616Qwen3.8 27BQwen27.8BScore72.66Score / $49Score / GB VRAM~4.1ScoredJun 25, 202617Claude Opus 4.8AnthropicScore72.03Score / $3ScoredJun 25, 202618Grok 4.6xAIScore71.87Score / $12ScoredJun 25, 202619GPT-5.6 SolOpenAIScore71.85Score / $14ScoredJun 25, 202620Grok 4.5xAIScore71.53Score / $12ScoredJun 25, 202621Kimi K3Moonshot AI2.8TScore71.36Score / $7Score / GB VRAM~0.041ScoredJun 25, 202622DeepSeek V4 Flash Vision ExpDeepSeek305BScore70.96Score / $110Score / GB VRAM~0.37ScoredJun 25, 202623GPT-5.5OpenAIScore70.73Score / $5ScoredJun 25, 202624Claude Sonnet 5.5AnthropicScore70.52Score / $7ScoredJun 25, 202625GPT-5.4OpenAIScore70.22Score / $9ScoredJun 25, 202626InklingThinking Machines952BScore70.1Score / $17Score / GB VRAM~0.12ScoredJun 25, 202627DeepSeek V4.1 FlashDeepSeek763BScore70.03Score / $175Score / GB VRAM~0.15ScoredJun 25, 202628Muse Spark 1.1MetaScore69.64Score / $16ScoredJun 25, 202629GLM 5.3Z.aiScore69.3Score / $61ScoredJun 25, 202630GPT-6 SolOpenAIScore68.57Score / $14ScoredJun 25, 202631Gemini 3.5 Flash LiteGoogleScore67.24Score / $54ScoredJun 25, 202632GPT-5.4 NanoOpenAIScore67.2Score / $108ScoredJun 25, 202633Claude Opus 5.5AnthropicScore67.05Score / $3ScoredJun 25, 202634Claude Opus 4.7AnthropicScore66.74Score / $3ScoredJun 25, 202635GPT-5.2-CodexOpenAIScore66.45Score / $5ScoredJun 25, 202636Grok Build 0.1xAIScore65.22Score / $33ScoredJun 25, 202637GPT-5.6 TerraOpenAIScore64.62Score / $11ScoredJun 25, 202638Kimi K2.6Moonshot AI1.1TScore64.36Score / $35Score / GB VRAM~0.096ScoredJun 25, 202639Claude Sonnet 5AnthropicScore63.86Score / $6ScoredJun 25, 202640Claude Opus 5AnthropicScore63.77Score / $3ScoredJun 25, 202641Claude Opus 4.6AnthropicScore63.31Score / $3ScoredJun 25, 202642Claude Sonnet 4.6AnthropicScore63.22Score / $4ScoredJun 25, 202643DeepSeek V4 FlashDeepSeek291BScore63.14Score / $754Score / GB VRAM~0.34ScoredJun 25, 202644Grok 4.3xAIScore62.75Score / $25ScoredJun 25, 202645Claude Opus 4.5AnthropicScore62.55Score / $3ScoredJun 25, 202646DeepSeek V4 ProDeepSeek1.6TScore62.35Score / $40Score / GB VRAM~0.062ScoredJun 25, 202647GLM 5.2Z.ai753BScore62.29Score / $68Score / GB VRAM~0.13ScoredJun 25, 202648GPT-5.2 ChatOpenAIScore61.77Score / $9ScoredJun 25, 202649Ox AlphastealthScore60.25ScoredJun 25, 202650GPT-5.6 LunaOpenAIScore60.12Score / $100ScoredJun 25, 202651GPT-5.4 MiniOpenAIScore59.8Score / $27ScoredJun 25, 202652Union AlphastealthScore59.53ScoredJun 25, 202653Qwen3.6 PlusQwenScore58.34Score / $30ScoredJun 25, 202654MiniMax M3MiniMax427BScore57.51Score / $60Score / GB VRAM~0.21ScoredJun 25, 202655Kimi K2.7 CodeMoonshot AI1.1TScore56.29Score / $19Score / GB VRAM~0.084ScoredJun 25, 202656GPT-6 LunaOpenAIScore55.93Score / $224ScoredJun 25, 202657Qwen3.6 27BQwen27.8BScore53.23Score / $27Score / GB VRAM~3ScoredJun 25, 202658GLM 5.3 FlashZ.ai321BScore52.82Score / $211Score / GB VRAM~0.26ScoredJun 25, 2026