| 1 | Claude Opus 5 | Anthropic | Proprietary | 96% | |
| 2 | Claude Mythos 5 | Anthropic | Proprietary | 95.5% | |
| 3 | Claude Fable 5 | Anthropic | Proprietary | 95% | |
| 4 | Claude Opus 4.8 | Anthropic · Closed | 88.6% | | |
| 5 | Claude Opus 4.7 (Adaptive) | Anthropic · Closed | 87.6% | | |
| 6 | Ornith-1.5-397B | Ornith AI · Open weight | 86% | | |
| 7 | Claude Sonnet 5 | Anthropic · Closed | 85.2% | | |
| 8 | GPT-5.3 Codex | OpenAI · Closed | 85% | | |
| 9 | Ornith-1.0-397B | DeepReinforce AI · Open weight | 82.4% | | |
| 10 | Claude Opus 4.5 | Anthropic · Closed | 80.9% | | |
| 11 | Claude Opus 4.6 | Anthropic · Closed | 80.8% | | |
| 12 | DeepSeek V4 Pro 0813 | DeepSeek · Closed | 80.6% | | |
| 13 | MiniMax M3 | MiniMax · Open weight | 80.5% | | |
| 14 | Qwen3.7 Max | Alibaba · Closed | 80.4% | | |
| 15 | Kimi K2.6 | Moonshot AI · Open weight | 80.2% | | |
| 16 | Inkling-Small | Thinking Machines Lab · Open weight | 80.2% | | |
| 17 | GPT-5.2 | OpenAI · Closed | 80% | | |
| 18 | Claude Sonnet 4.6 | Anthropic · Closed | 79.6% | | |
| 20 | DeepSeek V4 Flash 0731 | DeepSeek · Closed | 79% | | |
| 21 | Ornith-1.5-35B-A3B | Ornith AI · Open weight | 79% | | |
| 22 | Qwen3.6 Plus | Alibaba · Closed | 78.8% | | |
| 24 | dots3-note Preview | Dots Studio · Open weight | 78.4% | | |
| 25 | BTL-4 | Bad Theory Labs · Open weight | 78.4% | | |
| 26 | MiMo-V2-Pro | Xiaomi · Closed | 78% | | |
| 27 | GLM-5 | Z.AI · Open weight | 77.8% | | |
| 28 | Qwen3.7 Plus | Alibaba · Closed | 77.7% | | |
| 29 | Apodex 1.1 | Apodex · Closed | 77.7% | | |
| 30 | Inkling | Thinking Machines Lab · Open weight | 77.6% | | |
| 31 | Mistral Medium 3.5 128B | Mistral · Open weight | 77.6% | | |
| 32 | Muse Spark | Meta · Closed | 77.4% | | |
| 33 | Qwen3.6-27B | Alibaba · Open weight | 77.2% | | |
| 34 | Claude Sonnet 4.5 | Anthropic · Closed | 77.2% | | |
| 35 | Kimi K2.5 | Moonshot AI · Open weight | 76.8% | | |
| 36 | Kimi K2.5 (Reasoning) | Moonshot AI · Closed | 76.8% | | |
| 37 | Grok 4.20 | xAI · Closed | 76.7% | | |
| 38 | Qwen3.5 397B | Alibaba · Open weight | 76.2% | | |
| 39 | Muse Glimmer 30B | Meta · Open weight | 76% | | |
| 40 | Ornith-1.0-35B | DeepReinforce AI · Open weight | 75.6% | | |
| 41 | MiMo-V2-Omni | Xiaomi · Closed | 74.8% | | |
| 42 | Laguna M.1 | Poolside · Closed | 74.6% | | |
| 43 | Claude 4.1 Opus | Anthropic · Closed | 74.5% | | |
| 44 | Hy3 Preview | Tencent · Open weight | 74.4% | | |
| 45 | GLM-4.7 | Z.AI · Open weight | 73.8% | | |
| 48 | MAI-Thinking-1 | Microsoft · Closed | 73.5% | | |
| 49 | Qwen3.6-35B-A3B | Alibaba · Open weight | 73.4% | | |
| 50 | MiMo-V2-Flash | Xiaomi · Open weight | 73.4% | | |
| 51 | Claude Haiku 4.5 | Anthropic · Closed | 73.3% | | |
| 52 | Claude 4 Sonnet | Anthropic · Closed | 72.7% | | |
| 53 | MAI-Code-1.1-Flash | Microsoft · Closed | 72.6% | | |
| 54 | Qwen3.5-27B | Alibaba · Open weight | 72.4% | | |
| 55 | Qwen3.5-122B-A10B | Alibaba · Open weight | 72% | | |
| 56 | Nemotron 3 Ultra | NVIDIA · Open weight | 71.9% | | |
| 57 | Laguna XS 2.1 | Poolside · Open weight | 70.9% | | |
| 58 | Grok Code Fast 1 | xAI · Closed | 70.8% | | |
| 59 | Ornith-1.5-9B | Ornith AI · Open weight | 70.6% | | |
| 60 | Solar Pro 4 | Upstage · Closed | 70.6% | | |
| 61 | Solar Open 2 | Upstage · Open weight | 70.4% | | |
| 62 | Laguna XS.2 | Poolside · Open weight | 69.9% | | |
| 63 | Ornith-1.0-9B | DeepReinforce AI · Open weight | 69.4% | | |
| 64 | Qwen3.5-35B-A3B | Alibaba · Open weight | 69.2% | | |
| 65 | LongCat-Flash-Lite-Sparse | Meituan · Open weight | 68.2% | | |
| 66 | K-EXAONE 2.0 | LG AI Research · Open weight | 68.2% | | |
| 67 | Gemini 2.5 Pro | Google · Closed | 63.8% | | |
| 68 | Ternary Bonsai 2 27B | Prism ML · Open weight | 60.8% | | |
| 69 | Granite 4.2 30B | IBM · Open weight | 57% | | |
| 70 | GPT-4.1 | OpenAI · Closed | 54.6% | | |
| 71 | ZAYA1-74B-Preview | Zyphra · Open weight | 53.2% | | |
| 72 | Nemotron 3.5 Lightning 30B A3B NVFP4 | NVIDIA · Open weight | 52.8% | | |
| 73 | o3-mini | OpenAI · Closed | 49.3% | | |
| 74 | LLaDA2.2-flash | InclusionAI · Open weight | 49.3% | | |
| 75 | Claude 3.5 Sonnet | Anthropic · Closed | 49% | | |
| 76 | Granite 4.2 8B | IBM · Open weight | 47.7% | | |
| 77 | MiniCPM5-2B | OpenBMB · Open weight | 46.4% | | |
| 78 | DeepSeek V3 | DeepSeek · Open weight | 42% | | |
| 79 | GPT-4.1 mini | OpenAI · Closed | 23.6% | | |