Leaderboards
Benchmarks
Every score carries its origin.
Arena Hard Prompts168GPQA Diamond16LiveBench58LiveBench Data Analysis58LiveBench Reasoning58MMLU-Pro16
LiveBench Data Analysis
% average over LiveBench table and event-ordering tasksProvenance58 modelsscores Jun 25, 2026
1GPT-6 AstraOpenAIScore82.97Score / $3ScoredJun 25, 20262GPT-5.5OpenAIScore81.58Score / $5ScoredJun 25, 20263GPT-6 SolOpenAIScore81.19Score / $16ScoredJun 25, 20264Claude Fable 5AnthropicScore80.54Score / $2ScoredJun 25, 20265Claude Opus 5.5AnthropicScore80.31Score / $4ScoredJun 25, 20266Claude Fable 5.1AnthropicScore80.28Score / $2ScoredJun 25, 20267GPT-5.6 SolOpenAIScore79.84Score / $16ScoredJun 25, 20268Muse Spark 1.3MetaScore79.57Score / $19ScoredJun 25, 20269DeepSeek V4 Flash Vision ExpDeepSeek305BScore79.48Score / $123Score / GB VRAM~0.41ScoredJun 25, 202610GPT-5.4OpenAIScore79.31Score / $11ScoredJun 25, 202611GPT-5.6 TerraOpenAIScore79.31Score / $13ScoredJun 25, 202612DeepSeek V4.1 FlashDeepSeek763BScore79.25Score / $198Score / GB VRAM~0.16ScoredJun 25, 202613Kimi K3Moonshot AI2.8TScore78.73Score / $8Score / GB VRAM~0.045ScoredJun 25, 202614Claude Sonnet 5.5AnthropicScore78.62Score / $8ScoredJun 25, 202615Qwen3.8 MaxQwenScore78.41Score / $16ScoredJun 25, 202616Qwen3 MaxQwenScore78.41Score / $20ScoredJun 25, 202617Claude Opus 4.7AnthropicScore78.26Score / $3ScoredJun 25, 202618GPT-5.2-CodexOpenAIScore78.2Score / $6ScoredJun 25, 202619GPT-5.2 ChatOpenAIScore78.16Score / $11ScoredJun 25, 202620GPT-5.6 LunaOpenAIScore78.03Score / $130ScoredJun 25, 202621Claude Sonnet 4.6AnthropicScore77.95Score / $5ScoredJun 25, 202622Grok 4.7xAIScore76.88Score / $13ScoredJun 25, 202623Qwen3.8 27BQwen27.8BScore76.59Score / $51Score / GB VRAM~4.4ScoredJun 25, 202624Muse Spark 1.2MetaScore76.46Score / $18ScoredJun 25, 202625GLM 5.3 FlashZ.ai321BScore76.4Score / $306Score / GB VRAM~0.38ScoredJun 25, 202626MiniMax M3MiniMax427BScore76.17Score / $79Score / GB VRAM~0.28ScoredJun 25, 202627Ox AlphastealthScore75.77ScoredJun 25, 202628Union AlphastealthScore74.6ScoredJun 25, 202629Claude Opus 5AnthropicScore74.55Score / $3ScoredJun 25, 202630DeepSeek V4 ProDeepSeek1.6TScore74.54Score / $48Score / GB VRAM~0.074ScoredJun 25, 202631Claude Opus 4.5AnthropicScore74.44Score / $3ScoredJun 25, 202632Qwen3.8 FlashQwenScore74.24Score / $194ScoredJun 25, 202633Grok 4.6xAIScore73.86Score / $12ScoredJun 25, 202634GLM 5.2Z.ai753BScore73.74Score / $81Score / GB VRAM~0.16ScoredJun 25, 202635GPT-6 LunaOpenAIScore73.37Score / $293ScoredJun 25, 202636Grok 4.5xAIScore73.04Score / $12ScoredJun 25, 202637InklingThinking Machines952BScore72.78Score / $18Score / GB VRAM~0.12ScoredJun 25, 202638Muse Spark 1.1MetaScore72.55Score / $17ScoredJun 25, 202639Qwen3.7 MaxQwenScore71.79Score / $19ScoredJun 25, 202640Claude Sonnet 5AnthropicScore71.74Score / $7ScoredJun 25, 202641GPT-5.4 MiniOpenAIScore70.79Score / $31ScoredJun 25, 202642Grok Build 0.1xAIScore70.79Score / $35ScoredJun 25, 202643Qwen3.6 27BQwen27.8BScore70.43Score / $35Score / GB VRAM~4ScoredJun 25, 202644GLM 5.3Z.aiScore70.24Score / $62ScoredJun 25, 202645Qwen3.6 PlusQwenScore69.91Score / $36ScoredJun 25, 202646Claude Opus 4.6AnthropicScore69.89Score / $3ScoredJun 25, 202647DeepSeek V4 FlashDeepSeek291BScore68.02Score / $812Score / GB VRAM~0.37ScoredJun 25, 202648Gemini 3.7 FlashGoogleScore67.96Score / $36ScoredJun 25, 202649GPT-5.4 NanoOpenAIScore67.64Score / $108ScoredJun 25, 202650Claude Opus 4.8AnthropicScore66.03Score / $3ScoredJun 25, 202651Kimi K2.6Moonshot AI1.1TScore65.13Score / $36Score / GB VRAM~0.098ScoredJun 25, 202652Gemini 3.5 FlashGoogleScore64.86Score / $14ScoredJun 25, 202653Gemini 3.6 FlashGoogleScore63Score / $34ScoredJun 25, 202654Kimi K2.7 CodeMoonshot AI1.1TScore62.66Score / $21Score / GB VRAM~0.094ScoredJun 25, 202655Grok 4.3xAIScore55.77Score / $22ScoredJun 25, 202656Nemotron 3 UltraNVIDIA561BScore54.46Score / $25Score / GB VRAM~0.15ScoredJun 25, 202657Gemini 3.8 FlashGoogleScore54.01Score / $29ScoredJun 25, 202658Gemini 3.5 Flash LiteGoogleScore53.25Score / $43ScoredJun 25, 2026