| 1 | Claude Fable 5 | Anthropic | Proprietary | 88.56% | |
| 2 | Claude Fable 5.1 | Anthropic | Proprietary | 88.51% | |
| 3 | Google | high reasoning | Gemini 3.1 Pro Preview | 87.40% | |
| 4 | Gemini 3.7 Flash | Google · Closed | high reasoning | 87.26% | |
| 5 | Gemini 3 Pro Preview | Google | high reasoning | 87.03% | |
| 6 | Gemini 3.8 Flash | Google · Closed | high reasoning | 86.99% | |
| 7 | Claude Opus 5 | Anthropic · Closed | 86.97% | | |
| 8 | GPT-5.6 Sol | OpenAI · Closed | max reasoning | 86.97% | |
| 9 | Gemini 3 Flash Preview | Google | high reasoning | 86.86% | |
| 10 | Gemini 3.6 Flash | Google · Closed | high reasoning | 86.70% | |
| 11 | GPT-5.5 | OpenAI · Closed | xhigh reasoning | 86.52% | |
| 12 | Grok 4.6 | xAI · Closed | high reasoning | 86.31% | |
| 13 | GPT-5.4 | OpenAI · Closed | xhigh reasoning | 86.04% | |
| 14 | GPT-5 | OpenAI | high reasoning | 86.02% | |
| 15 | Kimi K3 | Moonshot AI · Closed | 86.02% | | |
| 16 | Grok 4.5 | xAI · Closed | high reasoning | 85.97% | |
| 17 | GPT-5.1 | OpenAI · Closed | high reasoning | 85.68% | |
| 18 | MiniMax M3 | MiniMax · Open weight | 85.42% | | |
| 19 | Claude Opus 4.6 (Adaptive) | Anthropic · Closed | 85.30% | | |
| 20 | Muse Spark 1.2 | Meta · Closed | xhigh reasoning | 85.26% | |
| 21 | Claude Opus 4.7 | Anthropic · Closed | 85.25% | | |
| 22 | GPT-5.6 Terra | OpenAI · Closed | max reasoning | 85.11% | |
| 23 | Qwen3.5 Plus Thinking | Alibaba | 85.10% | | |
| 24 | Muse Spark 1.1 | Meta · Closed | xhigh reasoning | 84.98% | |
| 25 | Qwen3.7 Max | Alibaba · Closed | 84.91% | | |
| 26 | GLM-5.3 | Z.AI · Open weight | max reasoning | 84.84% | |
| 27 | Kimi K2.6 | Moonshot AI · Open weight | 84.74% | | |
| 28 | Claude Opus 4.5 Thinking | Anthropic · Closed | 84.60% | | |
| 29 | Grok 4.3 | xAI · Closed | 84.46% | | |
| 30 | GLM-5.1 | Z.AI · Open weight | 84.39% | | |
| 31 | Grok 4.7 | xAI · Closed | xhigh reasoning | 84.39% | |
| 32 | Gemini 2.5 Pro Exp 03 25 | Google | 84.32% | | |
| 33 | Qwen3.5 Flash | Alibaba · Closed | 84.28% | | |
| 34 | Qwen3.6 Plus | Alibaba · Closed | 84.23% | | |
| 35 | Muse Spark | Meta · Closed | 84.22% | | |
| 36 | Claude Sonnet 4.5 Thinking | Anthropic · Closed | 84.08% | | |
| 37 | Gemini 3.5 Flash-Lite | Google · Closed | high reasoning | 84.07% | |
| 38 | GLM-5.2 | Z.AI · Open weight | 84.07% | | |
| 39 | GLM 5 Thinking | Zhipu AI | 84.06% | | |
| 40 | GPT-5.6 Luna | OpenAI · Closed | max reasoning | 84.03% | |
| 41 | MiniMax M2.7 | MiniMax · Open weight | 83.98% | | |
| 42 | GLM-5.3-Flash | Z.AI · Open weight | max reasoning | 83.93% | |
| 43 | Claude Sonnet 5 | Anthropic · Closed | 83.92% | | |
| 44 | Gemini 2.5 Flash Preview 04 17 | Google | 83.80% | | |
| 45 | Gemini 3.1 Flash Lite Preview | Google | high reasoning | 83.76% | |
| 46 | o3 | OpenAI · Closed | high reasoning | 83.76% | |
| 47 | Hy4 preview | Tencent · Open weight | 83.76% | | |
| 48 | Qwen3.8 Max | Alibaba · Open weight | 83.61% | | |
| 49 | Gemini 3.5 Flash | Google · Closed | high reasoning | 83.60% | |
| 50 | Claude Opus 4.8 | Anthropic · Closed | 83.57% | | |
| 51 | Claude Opus 4.1 | Anthropic | 83.46% | | |
| 52 | GLM-4.7 | Z.AI · Open weight | 83.36% | | |
| 53 | DeepSeek V4.1 Flash | DeepSeek · Open weight | high reasoning | 83.28% | |
| 54 | Grok 4 0709 | xAI | 83.19% | | |
| 55 | Grok 3 Mini Fast High Reasoning | xAI | high reasoning | 83.14% | |
| 56 | GPT-4.1 | OpenAI · Closed | high reasoning | 83.10% | |
| 57 | Claude Opus 4 | Anthropic | 83.07% | | |
| 58 | Mercury 2.5 | Inception · Closed | high reasoning | 83.06% | |
| 59 | Inkling-Small | Thinking Machines Lab · Open weight | 0.99 reasoning | 82.99% | |
| 60 | Claude Sonnet 4 | Anthropic | 82.95% | | |
| 61 | Inkling | Thinking Machines Lab · Open weight | 0.99 reasoning | 82.95% | |
| 62 | Claude Opus 4.5 | Anthropic · Closed | 82.84% | | |
| 63 | GPT-5.2 | OpenAI · Closed | xhigh reasoning | 82.76% | |
| 64 | Gemini 2.5 Flash Preview 09 2025 Thinking | Google | 82.63% | | |
| 65 | Grok 3 | xAI | 82.59% | | |
| 66 | Claude 3.7 Sonnet 20250219 Thinking | Anthropic | 82.52% | | |
| 67 | Gemini 2.5 Flash Preview 09 2025 | Google | 82.49% | | |
| 68 | Grok 4.1 Fast (Reasoning) | xAI · Closed | 82.45% | | |
| 69 | Qwen3.8-27B | Alibaba · Open weight | xhigh reasoning | 82.43% | |
| 70 | DeepSeek V4 Pro 0813 | DeepSeek · Closed | max reasoning | 82.36% | |
| 71 | GPT-4o | OpenAI · Closed | high reasoning | 82.21% | |
| 72 | Claude Sonnet 4.6 | Anthropic · Closed | 82.12% | | |
| 73 | Nemotron 3 Ultra 550b A55b | Nvidia | 82.07% | | |
| 74 | Claude Sonnet 4 20250514 Thinking | Anthropic | 82.06% | | |
| 75 | Gemini 2.5 Flash Lite Preview 09 2025 Thinking | Google | 82.04% | | |
| 76 | Grok 3 Mini Fast Low Reasoning | xAI | low reasoning | 81.92% | |
| 77 | Qwen3 Max | Alibaba · Closed | 81.86% | | |
| 78 | GPT-5 mini | OpenAI · Closed | high reasoning | 81.77% | |
| 79 | Moonshotai Kimi K2 Instruct | Together AI | 81.45% | | |
| 80 | Claude Haiku 4.5 Thinking | Anthropic · Closed | 81.24% | | |
| 81 | DeepSeek V3 | DeepSeek · Open weight | 80.76% | | |
| 82 | Grok 4 Fast (Reasoning) | xAI · Closed | 80.60% | | |
| 83 | GPT-4 Turbo | OpenAI · Closed | high reasoning | 80.46% | |
| 84 | o1 | OpenAI · Closed | high reasoning | 80.39% | |
| 85 | Qwen3 Max Preview | Alibaba | 80.33% | | |
| 87 | Kimi K2 Thinking | Moonshot AI | 80.20% | | |
| 88 | Qwen3 235b A22b | Fireworks AI | 80.18% | | |
| 90 | Claude 3.7 Sonnet | Anthropic | 80.00% | | |
| 91 | MiniMax M2.5 | MiniMax · Closed | 79.96% | | |
| 92 | Command A 03 2025 | Cohere | 79.70% | | |
| 93 | Ling 3.0 Flash 2607 | Ant | 79.69% | | |
| 94 | GLM-4.6 | Z.AI · Open weight | 79.61% | | |
| 95 | Qwen Qwen2.5 72B Instruct Turbo | Together AI | 79.40% | | |
| 96 | O4 Mini | OpenAI | high reasoning | 79.19% | |
| 97 | Mistral Large 2512 | Mistral AI | 79.14% | | |
| 98 | Grok 4.1 Fast Non Reasoning | xAI | 79.11% | | |
| 99 | Gemini 2.5 Flash Lite Preview 09 2025 | Google | 79.01% | | |
| 100 | MiMo-V2.5 | Xiaomi · Closed | 78.89% | | |
| 101 | Grok 4 Fast Non Reasoning | xAI | 78.40% | | |
| 102 | Gemini 2.0 Flash 001 | Google | 78.36% | | |
| 103 | GPT-4.1 mini | OpenAI · Closed | high reasoning | 78.04% | |
| 104 | GPT-5.4 nano | OpenAI · Closed | high reasoning | 77.92% | |
| 105 | Llama4 Maverick Instruct Basic | Fireworks AI | 77.81% | | |
| 106 | Grok 4.20 0309 Reasoning | xAI | 77.74% | | |
| 107 | DeepSeek V3 0324 | Fireworks AI | 77.73% | | |
| 108 | DeepSeek V4 Flash 0731 | DeepSeek · Closed | high reasoning | 77.71% | |
| 109 | Meta Llama Llama 3.3 70B Instruct Turbo | Together AI | 77.18% | | |
| 110 | MiMo-V2.5-Pro | Xiaomi · Closed | 77.13% | | |
| 111 | DeepSeek V3p2 Thinking | Fireworks AI | 76.08% | | |
| 112 | GPT-OSS 120B | OpenAI · Open weight | 75.94% | | |
| 113 | GLM-4.5 | Z.AI · Closed | 75.63% | | |
| 114 | MiniMax M2.1 | MiniMax | 75.45% | | |
| 115 | Laguna M.1 | Poolside · Closed | 75.14% | | |
| 116 | Jamba 1.5 Large | AI21 Labs | 74.16% | | |
| 117 | Meta Llama Llama 4 Scout 17B 16E Instruct | Together AI | 72.04% | | |
| 118 | Jamba Large 1.6 | AI21 Labs | 71.96% | | |
| 119 | o3-mini | OpenAI · Closed | high reasoning | 71.54% | |
| 120 | Laguna XS.2 | Poolside · Open weight | 71.03% | | |
| 121 | GPT-OSS 20B | OpenAI · Open weight | 70.85% | | |
| 122 | Claude 3.5 Haiku | Anthropic | 70.33% | | |
| 123 | Gemini 1.0 Pro 002 | Google | 70.25% | | |
| 124 | Jamba Mini 1.6 | AI21 Labs | 69.73% | | |
| 125 | Qwen Qwen2.5 7B Instruct Turbo | Together AI | 69.56% | | |
| 126 | Mistral Small 2503 | Mistral AI | 69.16% | | |
| 127 | Gemini 1.5 Pro 002 | Google | 69.08% | | |
| 128 | Command R Plus | Cohere | 68.29% | | |
| 129 | Google Gemma 2 9b It | Together AI | 68.14% | | |
| 130 | Google Gemma 2 27b It | Together AI | 67.98% | | |
| 131 | DeepSeek-R1 | DeepSeek · Open weight | 67.32% | | |
| 132 | Jamba 1.5 Mini | AI21 Labs | 66.62% | | |
| 133 | DeepSeek V3p2 | Fireworks AI | 65.96% | | |
| 134 | GPT-3.5 Turbo | OpenAI | high reasoning | 64.37% | |
| 135 | Gemini 1.5 Flash 002 | Google | 63.24% | | |
| 136 | Meta Llama Llama 2 70b Hf | Together AI | 62.43% | | |
| 137 | Mistral Medium 2505 | Mistral AI | 61.98% | | |
| 138 | GPT-4.1 nano | OpenAI · Closed | high reasoning | 61.06% | |
| 139 | Mistralai Mixtral 8x7B V0.1 | Together AI | 55.84% | | |
| 140 | Magistral Medium 2509 | Mistral AI | 54.77% | | |
| 141 | Mistralai Mistral 7B V0.1 | Together AI | 53.77% | | |
| 142 | Togethercomputer Llama 2 13b | Together AI | 53.70% | | |
| 143 | Togethercomputer Llama 2 7b | Together AI | 51.87% | | |
| 144 | GPT-5 nano | OpenAI · Closed | high reasoning | 50.13% | |
| 145 | Magistral Small 2509 | Mistral AI | 40.02% | | |
| 146 | Command R | Cohere | 32.97% | | |