| 1 | Claude Opus 5.5 | Anthropic | Proprietary | 91.43% | |
| 2 | Claude Fable 5.1 | Anthropic | Proprietary | 91.29% | |
| 3 | Claude Opus 5 | Anthropic | Proprietary | 90.98% | |
| 4 | Muse Spark 1.2 | Meta · Closed | 90.06% | | |
| 5 | Grok 4.7 | xAI · Closed | 89.38% | | |
| 6 | GLM-5.3-Flash | Z.AI · Open weight | 88.94% | | |
| 7 | Muse Spark 1.1 | Meta · Closed | 88.89% | | |
| 8 | GLM-5.3 | Z.AI · Open weight | 88.81% | | |
| 9 | Claude Fable 5 | Anthropic · Closed | 88.52% | | |
| 10 | GPT-5.1 | OpenAI · Closed | 88.09% | | |
| 11 | Kimi K3 | Moonshot AI · Closed | 87.96% | | |
| 12 | GPT-6 Astra | OpenAI · Closed | 87.91% | | |
| 13 | MiniMax M3 | MiniMax · Open weight | 87.25% | | |
| 14 | Grok 4.5 | xAI · Closed | 86.88% | | |
| 15 | GPT-5.5 | OpenAI · Closed | 86.87% | | |
| 16 | Claude Opus 4.6 | Anthropic · Closed | 86.74% | | |
| 17 | Grok 4.6 | xAI · Closed | 86.53% | | |
| 18 | Claude Opus 4.6 (Adaptive) | Anthropic · Closed | 86.13% | | |
| 19 | Muse Spark | Meta · Closed | 85.90% | | |
| 20 | Claude Opus 4.8 | Anthropic · Closed | 85.75% | | |
| 21 | DeepSeek V4.1 Flash | DeepSeek · Open weight | 85.50% | | |
| 22 | Inkling | Thinking Machines Lab · Open weight | 85.41% | | |
| 23 | Claude Opus 4.5 Thinking | Anthropic · Closed | 85.32% | | |
| 24 | GPT-5.6 Sol | OpenAI · Closed | 85.23% | | |
| 25 | Claude Haiku 4.5 Thinking | Anthropic · Closed | 85.23% | | |
| 26 | Qwen3.8 Max | Alibaba · Open weight | 84.95% | | |
| 27 | Claude Sonnet 4.5 | Anthropic · Closed | 84.52% | | |
| 28 | Gemini 3.8 Flash | Google · Closed | 84.50% | | |
| 29 | GPT-5.6 Luna | OpenAI · Closed | 84.39% | | |
| 30 | GPT-5.2 | OpenAI · Closed | 84.39% | | |
| 31 | Inkling-Small | Thinking Machines Lab · Open weight | 84.11% | | |
| 32 | Claude Sonnet 4.5 Thinking | Anthropic · Closed | 84.10% | | |
| 33 | Gemini 3.7 Flash | Google · Closed | 83.94% | | |
| 34 | Qwen3.8-27B | Alibaba · Open weight | 83.85% | | |
| 35 | MiMo-V2.5-Pro | Xiaomi · Closed | 83.73% | | |
| 36 | GPT-5 | OpenAI | 83.65% | | |
| 37 | Hy4 preview | Tencent · Open weight | 83.60% | | |
| 38 | GLM-5.2 | Z.AI · Open weight | 83.53% | | |
| 39 | Claude Opus 4.5 | Anthropic · Closed | 83.25% | | |
| 40 | Gemini 2.5 Flash Thinking | Google | 82.98% | | |
| 41 | Claude Opus 4.7 | Anthropic · Closed | 82.95% | | |
| 42 | Gemini 2.5 Flash | Google · Closed | 82.87% | | |
| 43 | GPT-5.6 Terra | OpenAI · Closed | 82.87% | | |
| 44 | Grok 4 Fast (Reasoning) | xAI · Closed | 81.63% | | |
| 45 | Ling 3.0 Flash 2607 | ant | 80.90% | | |
| 46 | MiniMax M2.1 | MiniMax | 80.78% | | |
| 47 | GPT-5 mini | OpenAI · Closed | 80.58% | | |
| 48 | DeepSeek V4 Flash 0731 | DeepSeek · Closed | 80.36% | | |
| 49 | DeepSeek V4 Pro 0813 | DeepSeek · Closed | 80.17% | | |
| 50 | MiniMax M2.7 | MiniMax · Open weight | 79.87% | | |
| 51 | Grok 4 Fast Non Reasoning | xAI | 79.72% | | |
| 52 | Gemini 3.6 Flash | Google · Closed | 79.66% | | |
| 53 | Qwen3.7 Max | Alibaba · Closed | 79.40% | | |
| 54 | Grok 4.1 Fast (Reasoning) | xAI · Closed | 78.73% | | |
| 55 | Gemini 2.5 Flash Preview 09 2025 Thinking | Google | 78.50% | | |
| 56 | Grok 4 0709 | xAI | 78.15% | | |
| 57 | Kimi K2.6 | Moonshot AI · Open weight | 78.15% | | |
| 58 | Gemini 2.5 Flash Preview 09 2025 | Google | 77.95% | | |
| 59 | GPT-5.4 | OpenAI · Closed | 77.55% | | |
| 60 | Grok 4.1 Fast Non Reasoning | xAI | 77.46% | | |
| 61 | Qwen3 VL Plus | Alibaba | 77.13% | | |
| 62 | GPT-5.4 nano | OpenAI · Closed | 77.09% | | |
| 63 | Qwen3.6 Plus | Alibaba · Closed | 76.96% | | |
| 64 | o3 | OpenAI · Closed | 76.65% | | |
| 65 | Gemini 3.5 Flash | Google · Closed | 76.57% | | |
| 66 | Kimi K2.5 Thinking | Moonshot AI | 76.44% | | |
| 67 | Gemini 3.1 Pro Preview | Google | 76.11% | | |
| 68 | Claude Sonnet 5 | Anthropic · Closed | 76.05% | | |
| 69 | Gemini 2.5 Flash Lite Preview 09 2025 | Google | 75.82% | | |
| 71 | Grok 4.3 | xAI · Closed | 74.40% | | |
| 72 | Claude Opus 4.1 20250805 Thinking | Anthropic | 73.90% | | |
| 73 | Gemini 2.5 Pro | Google · Closed | 73.55% | | |
| 74 | GPT-5 nano | OpenAI · Closed | 72.86% | | |
| 75 | Gemini 2.5 Flash Lite | Google | 72.83% | | |
| 76 | Qwen3 Max | Alibaba · Closed | 72.71% | | |
| 77 | Claude Sonnet 4 | Anthropic | 72.41% | | |
| 78 | GLM-5.1 | Z.AI · Open weight | 72.27% | | |
| 79 | MiMo-V2.5 | Xiaomi · Closed | 72.15% | | |
| 80 | Gemini 3 Pro Preview | Google | 72.04% | | |
| 81 | Claude Opus 4.1 | Anthropic | 71.75% | | |
| 82 | Gemini 3.5 Flash-Lite | Google · Closed | 70.89% | | |
| 83 | Qwen3.5 Flash | Alibaba · Closed | 70.62% | | |
| 84 | Gemini 3 Flash Preview | Google | 69.92% | | |
| 85 | Claude Sonnet 4 20250514 Thinking | Anthropic | 69.35% | | |
| 86 | O4 Mini | OpenAI | 69.14% | | |
| 87 | GLM-4.7 | Z.AI · Open weight | 68.63% | | |
| 88 | Mistral Medium 3.5 | Mistral | 67.73% | | |
| 89 | Gemini 2.5 Flash Lite Preview 09 2025 Thinking | Google | 66.88% | | |
| 90 | Laguna M.1 | Poolside · Closed | 65.91% | | |
| 91 | Gemini 3.1 Flash Lite Preview | Google | 63.90% | | |
| 92 | Grok 4.20 0309 Reasoning | xAI | 63.41% | | |
| 93 | Laguna XS.2 | Poolside · Open weight | 61.43% | | |
| 94 | Mercury 2.5 | Inception · Closed | 55.09% | | |
| 95 | Llama4 Maverick Instruct Basic | Fireworks | 54.22% | | |
| 96 | Meta Llama Llama 4 Scout 17B 16E Instruct | Together AI | 50.59% | | |
| 97 | Nemotron Lightning 3p5 30b A3b | Fireworks | 4.27% | | |