| 1 | GPT-6 Astra | OpenAI | Proprietary | 96% | |
| 2 | Sakana Fugu-Ultra | Sakana AI | Proprietary | 95.5% | |
| 3 | Sakana Fugu | Sakana AI | Proprietary | 95.5% | |
| 4 | GPT-5.6 Sol | OpenAI · Closed | 94.6% | | |
| 5 | Claude Opus 4.7 (Adaptive) | Anthropic · Closed | 94.2% | | |
| 6 | Claude Mythos 5 | Anthropic · Closed | 94.1% | | |
| 7 | Claude Opus 4.8 | Anthropic · Closed | 93.6% | | |
| 8 | GPT-5.5 | OpenAI · Closed | 93.6% | | |
| 9 | Kimi K3 | Moonshot AI · Closed | 93.5% | | |
| 10 | GPT-5.6 Terra | OpenAI · Closed | 92.9% | | |
| 11 | GPT-5.4 | OpenAI · Closed | 92.8% | | |
| 12 | Ornith-1.5-397B | Ornith AI · Open weight | 92.8% | | |
| 13 | Qwen3.8 Max | Alibaba · Open weight | 92.6% | | |
| 14 | Qwen3.7 Max | Alibaba · Closed | 92.4% | | |
| 15 | GPT-5.2 | OpenAI · Closed | 92.4% | | |
| 16 | GPT-5.6 Luna | OpenAI · Closed | 92.3% | | |
| 17 | Hy4 preview | Tencent · Open weight | 92.3% | | |
| 18 | Gemini 3.5 Flash | Google · Closed | 92.2% | | |
| 19 | Qwen3.8-Flash-Next | Alibaba · Open weight | 91.7% | | |
| 20 | Claude Opus 4.6 | Anthropic · Closed | 91.3% | | |
| 21 | GLM-5.2 | Z.AI · Open weight | 91.2% | | |
| 22 | Qwen3.8-Omni-Flash | Alibaba · Closed | 91% | | |
| 23 | DeepSeek V4.1 Flash | DeepSeek · Open weight | 90.9% | | |
| 24 | Kimi K2.6 | Moonshot AI · Open weight | 90.5% | | |
| 25 | Qwen3.6 Plus | Alibaba · Closed | 90.4% | | |
| 26 | Qwen3.7 Plus | Alibaba · Closed | 90.3% | | |
| 27 | DeepSeek V4 Pro 0813 | DeepSeek · Closed | 90.1% | | |
| 28 | Grok 4.3 | xAI · Closed | 90.1% | | |
| 29 | Claude Sonnet 4.6 | Anthropic · Closed | 89.9% | | |
| 30 | Interfaze Beta | Interfaze · Closed | 89.9% | | |
| 31 | Inkling-Small | Thinking Machines Lab · Open weight | 89.5% | | |
| 32 | Qwen3.8-27B | Alibaba · Open weight | 89.2% | | |
| 33 | Ornith-1.5-35B-A3B | Ornith AI · Open weight | 89.2% | | |
| 35 | Qwen3.5 397B | Alibaba · Open weight | 88.4% | | |
| 36 | DeepSeek V4 Flash 0731 | DeepSeek · Closed | 88.1% | | |
| 37 | GPT-5.4 mini | OpenAI · Closed | 88% | | |
| 38 | Inkling | Thinking Machines Lab · Open weight | 87.9% | | |
| 39 | Qwen3.6-27B | Alibaba · Open weight | 87.8% | | |
| 40 | Kimi K2.5 | Moonshot AI · Open weight | 87.6% | | |
| 41 | Kimi K2.5 (Reasoning) | Moonshot AI · Closed | 87.6% | | |
| 43 | Hy3 Preview | Tencent · Open weight | 87.2% | | |
| 44 | Claude Opus 4.5 | Anthropic · Closed | 87% | | |
| 45 | Nemotron 3 Ultra | NVIDIA · Open weight | 87% | | |
| 46 | Qwen3.5-122B-A10B | Alibaba · Open weight | 86.6% | | |
| 47 | Ornith-1.5-9B | Ornith AI · Open weight | 86.4% | | |
| 48 | GLM-5 | Z.AI · Open weight | 86% | | |
| 49 | Qwen3.6-35B-A3B | Alibaba · Open weight | 86% | | |
| 50 | Ternary Bonsai 2 27B | Prism ML · Open weight | 85.8% | | |
| 51 | GLM-4.7 | Z.AI · Open weight | 85.7% | | |
| 52 | Qwen3.5-27B | Alibaba · Open weight | 85.5% | | |
| 53 | Ling 3.0 Flash | InclusionAI · Open weight | 85.0% | | |
| 54 | Gemma 4 31B | Google · Open weight | 84.3% | | |
| 55 | Qwen3.5-35B-A3B | Alibaba · Open weight | 84.2% | | |
| 56 | MAI-Thinking-1 | Microsoft · Closed | 84.2% | | |
| 57 | Ling 3.0 Flash FP8 | InclusionAI · Open weight | 84% | | |
| 58 | MiMo-V2-Flash | Xiaomi · Open weight | 83.7% | | |
| 59 | Claude Sonnet 4.5 | Anthropic · Closed | 83.4% | | |
| 60 | Gemini 2.5 Pro | Google · Closed | 83% | | |
| 61 | GPT-5.4 nano | OpenAI · Closed | 82.8% | | |
| 62 | o1-pro | OpenAI · Closed | 79% | | |
| 63 | Gemma 4 12B | Google · Open weight | 78.8% | | |
| 64 | Qwen3 235B 2507 | Alibaba · Open weight | 77.5% | | |
| 65 | o3-mini | OpenAI · Closed | 77.2% | | |
| 66 | o1 | OpenAI · Closed | 75.7% | | |
| 67 | Nemotron 3.5 Lightning 30B A3B NVFP4 | NVIDIA · Open weight | 75.6% | | |
| 68 | Mercury 2 | Inception · Closed | 73.6% | | |
| 70 | Nemotron 3 Nano Omni 30B A3B | NVIDIA · Open weight | 72.2% | | |
| 72 | GPT-5 nano | OpenAI · Closed | 71.2% | | |
| 73 | ZAYA1-8B | Zyphra · Open weight | 71% | | |
| 74 | Granite 4.2 30B | IBM · Open weight | 66.4% | | |
| 75 | GPT-4.1 | OpenAI · Closed | 66.3% | | |
| 76 | GPT-4.1 mini | OpenAI · Closed | 64.2% | | |
| 77 | Granite 4.2 8B | IBM · Open weight | 64.1% | | |
| 78 | Claude 3.5 Sonnet | Anthropic · Closed | 59.4% | | |
| 79 | DeepSeek V3 | DeepSeek · Open weight | 59.1% | | |
| 80 | Ling 2.6 Flash | InclusionAI · Open weight | 59% | | |
| 81 | Gemma 4 E4B | Google · Open weight | 58.6% | | |
| 82 | Mellum2-12B-A2.5B-Thinking | JetBrains · Open weight | 57.6% | | |
| 83 | ZAYA1-74B-Preview | Zyphra · Open weight | 57.3% | | |
| 84 | Granite 4.2 3B | IBM · Open weight | 54.8% | | |
| 85 | GPT-4.1 nano | OpenAI · Closed | 50.3% | | |
| 86 | Soofi S 30B-A3B | Soofi Project · Open weight | 43.4% | | |
| 87 | Gemma 4 E2B | Google · Open weight | 43.4% | | |
| 88 | Mellum2-12B-A2.5B-Instruct | JetBrains · Open weight | 40.9% | | |
| 89 | LFM2.5-VL-450M | LiquidAI · Open weight | 25.7% | | |
| 90 | LFM2.5-230M | LiquidAI · Open weight | 25.4% | | |