rank,name,org,type,access,overall,categorization,vendor,context,coa,splits,tax,reliability,latency_s,cost_index 1,Jupid Agent,Jupid,Agent,,94.6,95.4,94,96.2,94.7,93,94,92,4.5,7.6 2,Claude Mythos 5,Anthropic,Model,,93.3,93.7,93.8,92.4,92.5,93.5,92.2,88,6.4,10.9 3,GPT-5.6 Sol Preview,OpenAI,Model,,93,93.3,93.5,91.8,92,93.8,92,86,6,12.2 4,Claude Fable 5,Anthropic,Model,,92.8,93.2,93.4,91.8,92,93,91.5,87,5.7,10.3 5,GPT-5.5 Pro,OpenAI,Model,,92.4,92.8,93,91.2,91.7,92.8,91.7,86,6.5,13.8 6,Claude Opus 4.8,Anthropic,Model,,91.9,92.2,92.8,90.8,91.2,92.1,90.8,85,5.2,9.5 7,GPT-5.5,OpenAI,Model,,91.5,91.9,92.4,90.5,91,91.9,90.9,84,4.4,8 8,Gemini 3.1 Pro Preview,Google,Model,,91.1,91.4,92,89.9,90.7,91.5,90.3,83,5,8.5 9,Claude Opus 4.7,Anthropic,Model,,90.3,90.8,91.5,89.1,89.8,90.6,89.5,82,5.1,9 10,DeepSeek V4 Pro Preview,DeepSeek,Model,,89.7,90.2,90.9,88.4,89.3,90.1,88.9,81,4.8,5.9 11,Grok 4.3,xAI,Model,,89.1,89.6,90.4,87.9,88.7,89.6,88.3,80,4.7,6.8 12,Qwen3.7-Max,Alibaba,Model,,88.7,89.2,90,87.4,88.3,89.1,87.8,79,4.2,5.6 13,Gemini 3.5 Flash,Google,Model,,88.1,88.5,89.2,86.8,87.6,88.4,87.1,77,2.4,3 14,Claude Sonnet 4.6,Anthropic,Model,,87.8,88.2,89,86.7,87.4,88.1,87,77,3.4,5.2 15,GPT-5.5 Instant,OpenAI,Model,,87.3,87.7,88.5,86,86.8,87.6,86.4,76,1.9,2.4 16,Kimi K2.6,Moonshot AI,Model,,86.8,87.2,88,85.5,86.4,87,85.9,75,4,4.8 17,DeepSeek V4 Flash,DeepSeek,Model,,86.3,86.8,87.6,85,85.9,86.6,85.4,74,3.1,3.4 18,Qwen3.7-Plus,Alibaba,Model,,85.9,86.4,87,84.6,85.4,86.2,85,74,3.4,3.7 19,Grok Build 0.1,xAI,Model,,85.4,85.9,86.4,84.1,84.9,85.7,84.4,73,2.7,3.3 20,Mistral Medium 3.5,Mistral,Model,,85,85.5,86.1,83.7,84.6,85.3,84.1,72,3.6,4.3 21,Llama 4 Maverick,Meta,Model,,84.6,85.1,85.6,83.3,84.2,84.9,83.8,71,3.7,4 22,Mistral Large 3,Mistral,Model,,84.2,84.7,85.3,82.9,83.8,84.5,83.4,70,4.1,4.7 23,Gemini 3 Flash,Google,Model,,83.8,84.2,84.9,82.5,83.4,84.1,83,69,2.2,2.6 24,Claude Opus 4.6,Anthropic,Model,,83.5,84,84.7,82.2,83,83.8,82.7,69,4.8,8.6 25,Cohere Command A+,Cohere,Model,,83.1,83.5,84.1,81.7,82.6,83.3,82.2,68,3.7,4.1 26,Amazon Nova Premier,Amazon,Model,,82.7,83.1,83.7,81.3,82.1,82.9,81.8,67,3.4,3.8 27,Claude Haiku 4.5,Anthropic,Model,,82.3,82.7,83.4,81,81.8,82.5,81.5,66,2.1,2.5 28,Perplexity Sonar Reasoning Pro,Perplexity,Model,,82,82.4,83,80.6,81.4,82.1,81.1,65,4.5,4.1 29,Mistral Small 4,Mistral,Model,,81.5,81.9,82.5,80.2,81,81.7,80.8,64,2.7,2.8 30,Gemini 3.1 Flash-Lite,Google,Model,,81.1,81.5,82,79.8,80.6,81.2,80.4,63,1.7,1.8