| 1 | Claude Opus 4.8 | Anthropic | Proprietary | 95% | |
| 2 | Qwen3.8 Max | Alibaba | Open weights | 95% | |
| 4 | Claude Sonnet 4.6 | Anthropic · Closed | 91% | | |
| 5 | Claude Opus 4.5 | Anthropic · Closed | 90% | | |
| 7 | Claude Opus 4.6 | Anthropic · Closed | 87% | | |
| 9 | Claude Opus 4.7 | Anthropic · Closed | 83% | | |
| 10 | Claude Sonnet 5 | Anthropic · Closed | 80% | | |
| 11 | Claude Sonnet 4.5 | Anthropic · Closed | 79% | | |
| 13 | Qwen3.8-27B | Alibaba · Open weight | 79% | | |
| 15 | Qwen3.5 397B (Reasoning) | Alibaba · Open weight | 78% | | |
| 16 | Claude Fable 5.1 | Anthropic · Closed | 77% | | |
| 17 | Claude Haiku 4.5 | Anthropic · Closed | 77% | | |
| 18 | Kimi K3 | Moonshot AI · Closed | 74% | | |
| 21 | Claude Opus 5 | Anthropic · Closed | 73% | | |
| 23 | GLM-5.3 | Z.AI · Open weight | 72% | | |
| 24 | Qwen3.7 Max | Alibaba · Closed | 72% | | |
| 25 | Qwen3.6 Plus | Alibaba · Closed | 72% | | |
| 28 | GPT-6 Astra | OpenAI · Closed | 69% | | |
| 29 | Qwen3.5 397B | Alibaba · Open weight | 69% | | |
| 30 | Grok 4.20 Multi-agent | xAI · Closed | 67% | | |
| 32 | Gemini 3.5 Flash-Lite | Google · Closed | 66% | | |
| 33 | Grok 4.6 | xAI · Closed | 66% | | |
| 34 | Kimi K2.6 | Moonshot AI · Open weight | 65% | | |
| 35 | stealth/ox-alpha (low) | Stealth | 65% | | |
| 38 | MiniMax M3 | MiniMax · Open weight | 64% | | |
| 42 | MiMo-V2.5-Pro | Xiaomi · Closed | 62% | | |
| 45 | DeepSeek V4.1 Flash | DeepSeek · Open weight | 57% | | |
| 48 | Claude Fable 5 | Anthropic · Closed | 56% | | |
| 50 | Grok 4.20 | xAI · Closed | 56% | | |
| 51 | Grok 4.5 | xAI · Closed | 55% | | |
| 53 | GPT-5.6 Terra | OpenAI · Closed | 54% | | |
| 56 | Nemotron 3 Super 120B A12B | NVIDIA · Open weight | 54% | | |
| 57 | Kimi K2.5 | Moonshot AI · Open weight | 52% | | |
| 58 | Grok 4.3 | xAI · Closed | 50% | | |
| 60 | Muse Spark 1.2 | Meta · Closed | 50% | | |
| 65 | GPT-5.6 Sol | OpenAI · Closed | 48% | | |
| 66 | GPT-5.4 | OpenAI · Closed | 48% | | |
| 67 | Gemini 3 Pro | Google · Closed | 48% | | |
| 69 | GPT-5.5 | OpenAI · Closed | 47% | | |
| 74 | Claude 3.5 Sonnet | Anthropic · Closed | 46% | | |
| 78 | GPT-5.2-Codex | OpenAI · Closed | 45% | | |
| 79 | GPT-5.1 | OpenAI · Closed | 45% | | |
| 81 | Claude 4.1 Opus | Anthropic · Closed | 43% | | |
| 86 | GPT-5.6 Luna | OpenAI · Closed | 41% | | |
| 87 | GPT-5.3 Instant | OpenAI · Closed | 40% | | |
| 89 | DeepSeek V4 Flash 0731 | DeepSeek · Closed | 39% | | |
| 92 | Gemini 3.6 Flash | Google · Closed | 39% | | |
| 93 | GPT-5.2 | OpenAI · Closed | 38% | | |
| 96 | Gemini 3.1 Pro | Google · Closed | 37% | | |
| 99 | GPT-5.5 Pro | OpenAI · Closed | 36% | | |
| 100 | Gemini 3 Pro Deep Think | Google · Closed | 36% | | |
| 101 | DeepSeek V4 Pro 0813 | DeepSeek · Closed | 35% | | |
| 102 | Gemini 3.7 Flash | Google · Closed | 35% | | |
| 103 | MiMo-V2.5 | Xiaomi · Closed | 35% | | |
| 108 | GPT-5.4 mini | OpenAI · Closed | 32% | | |
| 111 | GPT-5.1-Codex-Max | OpenAI · Closed | 32% | | |
| 113 | Kimi K2.5 (Reasoning) | Moonshot AI · Closed | 31% | | |
| 116 | GLM-5-Turbo | Z.AI · Closed | 31% | | |
| 118 | GLM-5.2 | Z.AI · Open weight | 31% | | |
| 119 | Claude 4 Sonnet | Anthropic · Closed | 30% | | |
| 122 | Llama 4 Maverick | Meta · Open weight | 29% | | |
| 125 | GLM-5 (Reasoning) | Z.AI · Open weight | 28% | | |
| 128 | GPT-5.2 Instant | OpenAI · Closed | 27% | | |
| 129 | o3 | OpenAI · Closed | 26% | | |
| 132 | Gemma 4 31B | Google · Open weight | 25% | | |
| 133 | GPT-5.3 Codex | OpenAI · Closed | 24% | | |
| 136 | GLM-5.1 | Z.AI · Open weight | 22% | | |
| 137 | Step 3.5 Flash | StepFun · Open weight | 22% | | |
| 138 | Laguna S 2.1 | Poolside · Open weight | 22% | | |
| 140 | Gemma 4 26B A4B | Google · Open weight | 21% | | |
| 145 | Gemini 2.5 Pro | Google · Closed | 20% | | |
| 146 | GLM-5 | Z.AI · Open weight | 20% | | |
| 147 | Llama 4 Scout | Meta · Open weight | 20% | | |
| 149 | Gemini 3.5 Flash | Google · Closed | 20% | | |
| 151 | Grok 4.1 Fast | xAI · Closed | 19% | | |
| 152 | Gemini 2.5 Flash | Google · Closed | 19% | | |
| 158 | Trinity-Large-Thinking | Arcee AI · Open weight | 17% | | |
| 159 | MiMo-V2-Flash | Xiaomi · Open weight | 16% | | |
| 160 | Hy3 | Tencent · Open weight | 16% | | |
| 164 | GPT-5.4 nano | OpenAI · Closed | 14% | | |
| 167 | GPT-4.1 | OpenAI · Closed | 14% | | |
| 169 | DeepSeek V3.2 (Thinking) | DeepSeek · Open weight | 13% | | |
| 173 | GPT-4o | OpenAI · Closed | 12% | | |
| 174 | GPT-OSS 120B | OpenAI · Open weight | 12% | | |
| 176 | Gemini 3.1 Flash-Lite | Google · Closed | 11% | | |
| 177 | Seed 1.6 | ByteDance · Closed | 11% | | |
| 180 | Gemini 3 Flash | Google · Closed | 10% | | |
| 181 | DeepSeek V3.2 | DeepSeek · Open weight | 10% | | |
| 182 | Claude 3 Haiku | Anthropic · Closed | 10% | | |
| 185 | Kimi K2 | Moonshot AI · Closed | 10% | | |
| 187 | MiniMax M2.5 | MiniMax · Closed | 9% | | |
| 190 | GLM-4.5 | Z.AI · Closed | 8% | | |
| 191 | MiniMax M2.7 | MiniMax · Open weight | 8% | | |
| 192 | DeepSeek-R1 | DeepSeek · Open weight | 8% | | |
| 193 | o4-mini (high) | OpenAI · Closed | 8% | | |
| 214 | GPT-4o mini | OpenAI · Closed | 2% | | |