Leaderboards
Benchmarks
Every score carries its origin.
LiveBench Mathematics
% average over LiveBench competition and olympiad maths tasksProvenance58 modelsscores Jun 25, 2026
1Claude Opus 5.5AnthropicScore97.08Score / $5ScoredJun 25, 20262Claude Fable 5.1AnthropicScore97.01Score / $2ScoredJun 25, 20263GPT-6 AstraOpenAIScore96.81Score / $4ScoredJun 25, 20264Claude Sonnet 5.5AnthropicScore96.69Score / $10ScoredJun 25, 20265GPT-6 SolOpenAIScore96.36Score / $19ScoredJun 25, 20266GPT-5.6 SolOpenAIScore96.2Score / $19ScoredJun 25, 20267Claude Fable 5AnthropicScore95.99Score / $2ScoredJun 25, 20268Muse Spark 1.3MetaScore95.95Score / $23ScoredJun 25, 20269GPT-5.5OpenAIScore95.86Score / $6ScoredJun 25, 202610Claude Opus 5AnthropicScore95.73Score / $4ScoredJun 25, 202611Grok 4.7xAIScore95.68Score / $16ScoredJun 25, 202612Union AlphastealthScore95.29ScoredJun 25, 202613GPT-5.6 TerraOpenAIScore94.91Score / $16ScoredJun 25, 202614Claude Opus 4.8AnthropicScore94.32Score / $4ScoredJun 25, 202615GPT-5.4OpenAIScore94.15Score / $13ScoredJun 25, 202616Gemini 3.7 FlashGoogleScore93.47Score / $50ScoredJun 25, 202617DeepSeek V4.1 FlashDeepSeek763BScore93.29Score / $233Score / GB VRAM~0.19ScoredJun 25, 202618GPT-5.2 ChatOpenAIScore93.17Score / $13ScoredJun 25, 202619Claude Sonnet 5AnthropicScore92.94Score / $9ScoredJun 25, 202620Claude Opus 4.7AnthropicScore92.85Score / $4ScoredJun 25, 202621Grok 4.6xAIScore92.57Score / $15ScoredJun 25, 202622Gemini 3.8 FlashGoogleScore91.56Score / $49ScoredJun 25, 202623Qwen3 MaxQwenScore91.31Score / $23ScoredJun 25, 202624Qwen3.8 MaxQwenScore91.31Score / $18ScoredJun 25, 202625Muse Spark 1.2MetaScore91.2Score / $21ScoredJun 25, 202626GPT-5.4 NanoOpenAIScore90.98Score / $146ScoredJun 25, 202627Grok 4.5xAIScore90.82Score / $15ScoredJun 25, 202628DeepSeek V4 ProDeepSeek1.6TScore90.68Score / $58Score / GB VRAM~0.09ScoredJun 25, 202629Claude Opus 4.5AnthropicScore90.39Score / $4ScoredJun 25, 202630GLM 5.2Z.ai753BScore89.78Score / $99Score / GB VRAM~0.19ScoredJun 25, 202631Claude Opus 4.6AnthropicScore89.32Score / $4ScoredJun 25, 202632GPT-6 LunaOpenAIScore89.12Score / $356ScoredJun 25, 202633GPT-5.2-CodexOpenAIScore88.77Score / $6ScoredJun 25, 202634Nemotron 3 UltraNVIDIA561BScore88.66Score / $40Score / GB VRAM~0.25ScoredJun 25, 202635InklingThinking Machines952BScore88.36Score / $22Score / GB VRAM~0.15ScoredJun 25, 202636Gemini 3.5 FlashGoogleScore88.24Score / $20ScoredJun 25, 202637GLM 5.3Z.aiScore87.9Score / $77ScoredJun 25, 202638DeepSeek V4 Flash Vision ExpDeepSeek305BScore87.81Score / $136Score / GB VRAM~0.46ScoredJun 25, 202639GPT-5.6 LunaOpenAIScore87.2Score / $145ScoredJun 25, 202640Muse Spark 1.1MetaScore87.14Score / $21ScoredJun 25, 202641Claude Sonnet 4.6AnthropicScore86.99Score / $6ScoredJun 25, 202642Gemini 3.6 FlashGoogleScore86.4Score / $46ScoredJun 25, 202643Qwen3.8 27BQwen27.8BScore86.21Score / $58Score / GB VRAM~4.9ScoredJun 25, 202644Qwen3.8 FlashQwenScore85.82Score / $225ScoredJun 25, 202645Qwen3.7 MaxQwenScore85.25Score / $23ScoredJun 25, 202646Kimi K3Moonshot AI2.8TScore84.44Score / $9Score / GB VRAM~0.048ScoredJun 25, 202647Grok 4.3xAIScore84.34Score / $34ScoredJun 25, 202648Kimi K2.6Moonshot AI1.1TScore84.28Score / $46Score / GB VRAM~0.13ScoredJun 25, 202649Qwen3.6 PlusQwenScore83.72Score / $43ScoredJun 25, 202650GLM 5.3 FlashZ.ai321BScore81.24Score / $325Score / GB VRAM~0.4ScoredJun 25, 202651Qwen3.6 27BQwen27.8BScore79.87Score / $40Score / GB VRAM~4.6ScoredJun 25, 202652DeepSeek V4 FlashDeepSeek291BScore79.65Score / $951Score / GB VRAM~0.43ScoredJun 25, 202653Kimi K2.7 CodeMoonshot AI1.1TScore79.6Score / $27Score / GB VRAM~0.12ScoredJun 25, 202654GPT-5.4 MiniOpenAIScore78.46Score / $35ScoredJun 25, 202655Grok Build 0.1xAIScore78.43Score / $39ScoredJun 25, 202656Ox AlphastealthScore77.53ScoredJun 25, 202657MiniMax M3MiniMax427BScore76.95Score / $80Score / GB VRAM~0.29ScoredJun 25, 202658Gemini 3.5 Flash LiteGoogleScore73.74Score / $59ScoredJun 25, 2026