| 1 | Claude Fable 5.1 | Anthropic | Proprietary | 58.0% | |
| 2 | Claude Opus 5 | Anthropic | Proprietary | 56.2% | |
| 3 | Muse Spark 1.3 | Meta | Proprietary | 55.7% | |
| 4 | GLM-5.3 | Z.AI · Open weight | 53.4% | | |
| 5 | Grok 4.6 | xAI · Closed | 53.4% | | |
| 6 | GPT-6 Astra | OpenAI · Closed | 51.5% | | |
| 7 | Claude Fable 5 | Anthropic · Closed | 51.0% | | |
| 8 | Kimi K3 | Moonshot AI · Closed | 50.6% | | |
| 9 | GPT-5.6 Sol | OpenAI · Closed | 50.5% | | |
| 10 | Qwen3.8 Max Preview | Alibaba · Closed | 49.6% | | |
| 11 | DeepSeek V4 Pro 0813 | DeepSeek · Closed | 49.6% | | |
| 12 | Qwen3.8-27B | Alibaba · Open weight | 46.5% | | |
| 13 | Claude Sonnet 5 | Anthropic · Closed | 44.3% | | |
| 14 | Muse Spark 1.2 | Meta · Closed | 44.0% | | |
| 15 | GPT-5.6 Terra | OpenAI · Closed | 43.7% | | |
| 16 | GPT-5.6 Luna | OpenAI · Closed | 42.7% | | |
| 17 | Claude Opus 4.8 | Anthropic · Closed | 42.6% | | |
| 18 | Grok 4.5 | xAI · Closed | 42.1% | | |
| 19 | DeepSeek V4 Flash 0731 | DeepSeek · Closed | 41.7% | | |
| 20 | Gemini 3.8 Flash | Google · Closed | 41.1% | | |
| 21 | Claude Opus 4.7 (Adaptive) | Anthropic · Closed | 39.5% | | |
| 22 | GLM-5.2 | Z.AI · Open weight | 39.4% | | |
| 23 | GPT-5.5 | OpenAI · Closed | 37.3% | | |
| 24 | Gemini 3.7 Flash | Google · Closed | 36.4% | | |
| 26 | Quasar 438B | Multiverse Computing · Closed | 32.7% | | |
| 27 | MiniMax M3 | MiniMax · Open weight | 30.8% | | |
| 28 | Gemini 3.6 Flash | Google · Closed | 30.1% | | |
| 29 | Muse Spark 1.1 | Meta · Closed | 27.5% | | |
| 30 | Gemini 3.5 Flash | Google · Closed | 27.3% | | |
| 31 | Hy3 | Tencent · Open weight | 25.6% | | |
| 32 | Hy3 Preview | Tencent · Open weight | 25.6% | | |
| 33 | GLM-5.1 | Z.AI · Open weight | 25.2% | | |
| 34 | Inkling-Small | Thinking Machines Lab · Open weight | 24.9% | | |
| 35 | Inkling | Thinking Machines Lab · Open weight | 24.3% | | |
| 36 | Qwen3.7 Max | Alibaba · Closed | 23.9% | | |
| 37 | MiMo-V2.5-Pro | Xiaomi · Closed | 22.7% | | |
| 38 | Kimi K2.7 Code | Moonshot AI · Open weight | 22.5% | | |
| 39 | Kimi K2.6 | Moonshot AI · Open weight | 22.1% | | |
| 40 | Nemotron 3 Ultra | NVIDIA · Open weight | 21.7% | | |
| 41 | Ling 3.0 Flash | InclusionAI · Open weight | 21.0% | | |
| 42 | Ling 3.0 Flash FP8 | InclusionAI · Open weight | 21.0% | | |
| 43 | Qwen3.6-27B | Alibaba · Open weight | 20.1% | | |
| 44 | Qwen3.7 Plus | Alibaba · Closed | 19.7% | | |
| 45 | GPT-5.4 mini | OpenAI · Closed | 19.6% | | |
| 46 | GPT-5.4 nano | OpenAI · Closed | 17.7% | | |
| 47 | Grok 4.3 | xAI · Closed | 17.2% | | |
| 48 | MiniMax M2.7 | MiniMax · Open weight | 16.8% | | |
| 49 | Gemini 3.5 Flash-Lite | Google · Closed | 15.9% | | |
| 50 | Qwen3.6-35B-A3B | Alibaba · Open weight | 15.0% | | |
| 51 | Muse Glimmer 30B | Meta · Open weight | 10.5% | | |
| 52 | Gemini 3.1 Pro | Google · Closed | 10.3% | | |
| 53 | Qwen3.5-122B-A10B | Alibaba · Open weight | 9.6% | | |
| 54 | Mistral Medium 3.5 128B | Mistral · Open weight | 9.3% | | |
| 55 | GPT-5 mini | OpenAI · Closed | 8.9% | | |
| 56 | Gemma 4 31B | Google · Open weight | 6.7% | | |
| 57 | GPT-OSS 120B | OpenAI · Open weight | 6.2% | | |
| 58 | Nemotron 3.5 Lightning 30B A3B NVFP4 | NVIDIA · Open weight | 6.1% | | |
| 59 | Nemotron 3 Super 100B | NVIDIA · Open weight | 4.1% | | |
| 60 | Nemotron 3 Super 120B A12B | NVIDIA · Open weight | 4.1% | | |
| 61 | Mercury 2 | Inception · Closed | 4.0% | | |
| 62 | Granite 4.2 8B | IBM · Open weight | 3.7% | | |
| 63 | Command A+ | Cohere · Open weight | 3.6% | | |
| 64 | Gemini 2.5 Pro | Google · Closed | 3.5% | | |
| 65 | Mistral Large 3 | Mistral · Closed | 2.4% | | |
| 66 | Mistral Small 4 | Mistral · Open weight | 1.4% | | |
| 67 | Mistral Small 4 (Reasoning) | Mistral · Open weight | 1.4% | | |
| 68 | GPT-OSS 20B | OpenAI · Open weight | 1.4% | | |
| 69 | Trinity-Large-Thinking | Arcee AI · Open weight | 1.2% | | |
| 70 | Trinity-Large-Preview | Arcee AI · Open weight | 1.2% | | |
| 71 | Ministral 3 14B (Reasoning) | Mistral · Open weight | 1.1% | | |
| 72 | Ministral 3 14B | Mistral · Open weight | 1.1% | | |
| 73 | Nemotron 3 Nano 30B | NVIDIA · Open weight | 1.0% | | |
| 74 | DeepSeek V3 | DeepSeek · Open weight | 0.8% | | |
| 75 | Ministral 3 3B (Reasoning) | Mistral · Open weight | 0.8% | | |
| 76 | Ministral 3 3B | Mistral · Open weight | 0.8% | | |
| 77 | Celeris-1 | Celeris · Closed | 0.7% | | |
| 78 | Llama 4 Maverick | Meta · Open weight | 0.6% | | |
| 79 | Ministral 3 8B (Reasoning) | Mistral · Open weight | 0.6% | | |
| 80 | Ministral 3 8B | Mistral · Open weight | 0.6% | | |
| 81 | Llama 4 Scout | Meta · Open weight | 0.6% | | |
| 82 | Gemma 3 27B | Google · Open weight | 0.1% | | |