| 1 | — | OGPT-5.2 (xhigh)闭源OpenAI | 50.6 | — | — | — |
| 2 | — | AClaude Opus 4.5 (Reasoning)闭源Anthropic | 49.1 | — | — | — |
| 3 | — | GGemini 3 Pro Preview (high)闭源Google | 47.9 | — | — | — |
| 4 | — | OGPT-5.1 (high)闭源OpenAI | 47.0 | — | — | — |
| 5 | — | GGemini 3 Flash Preview (Reasoning)闭源Google | 45.8 | — | — | — |
| 6 | — | OGPT-5.2 (medium)闭源OpenAI | 45.4 | — | — | — |
| 7 | — | OGPT-5 (high)闭源OpenAI | 44.1 | — | — | — |
| 8 | — | OGPT-5 Codex (high)闭源OpenAI | 44.0 | — | — | — |
| 9 | — | AClaude Opus 4.5 (Non-reasoning)闭源Anthropic | 42.5 | — | — | — |
| 10 | — | AClaude 4.5 Sonnet (Reasoning)闭源Anthropic | 42.4 | — | — | — |
| 11 | — | OGPT-5 (medium)闭源OpenAI | 41.9 | — | — | — |
| 12 | — | ZGLM-4.7 (Reasoning)开源Z AI | 41.7 | — | — | — |
| 13 | — | OGPT-5.1 Codex (high)闭源OpenAI | 41.6 | — | — | — |
| 14 | — | XGrok 4闭源xAI | 41.4 | — | — | — |
| 15 | — | DDeepSeek V3.2 (Reasoning)开源DeepSeek | 41.2 | — | — | — |
| 16 | — | OGPT-5 mini (high)闭源OpenAI | 40.6 | — | — | — |
| 17 | — | GGemini 3 Pro Preview (low)闭源Google | 40.6 | — | — | — |
| 18 | — | KKimi K2 Thinking开源Kimi | 40.3 | — | — | — |
| 19 | — | MMiniMax-M2.1开源MiniMax | 39.4 | — | — | — |
| 20 | — | XMiMo-V2-Flash (Reasoning)开源Xiaomi | 39.1 | — | — | — |
| 21 | — | OGPT-5 (low)闭源OpenAI | 38.7 | — | — | — |
| 22 | — | OGPT-5 mini (medium)闭源OpenAI | 38.6 | — | — | — |
| 23 | — | AClaude 4 Sonnet (Reasoning)闭源Anthropic | 38.3 | — | — | — |
| 24 | — | XGrok 4.1 Fast (Reasoning)闭源xAI | 38.2 | — | — | — |
| 25 | — | OGPT-5.1 Codex mini (high)闭源OpenAI | 38.0 | — | — | — |
| 26 | — | AClaude 4.5 Haiku (Reasoning)闭源Anthropic | 36.6 | — | — | — |
| 27 | — | AClaude 4.5 Sonnet (Non-reasoning)闭源Anthropic | 36.6 | — | — | — |
| 28 | — | KKAT-Coder-Pro V1闭源KwaiKAT | 36.1 | — | — | — |
| 29 | — | MMiniMax-M2开源MiniMax | 35.7 | — | — | — |
| 30 | — | ANova 2.0 Pro Preview (medium)闭源Amazon | 35.3 | — | — | — |
| 31 | — | GGemini 3 Flash Preview (Non-reasoning)闭源Google | 34.8 | — | — | — |
| 32 | — | XGrok 4 Fast (Reasoning)闭源xAI | 34.6 | — | — | — |
| 33 | — | AClaude 3.7 Sonnet (Reasoning)闭源Anthropic | 34.4 | — | — | — |
| 34 | — | GGemini 2.5 Pro闭源Google | 34.1 | — | — | — |
| 35 | — | ZGLM-4.7 (Non-reasoning)开源Z AI | 33.8 | — | — | — |
| 36 | — | DDeepSeek V3.1 Terminus (Reasoning)开源DeepSeek | 33.5 | — | — | — |
| 37 | — | BDoubao Seed Code闭源ByteDance Seed | 33.2 | — | — | — |
| 38 | — | OGPT-5.2 (Non-reasoning)闭源OpenAI | 33.1 | — | — | — |
| 39 | — | Oo4-mini (high)闭源OpenAI | 32.9 | — | — | — |
| 40 | — | Ogpt-oss-120B (high)开源OpenAI | 32.9 | — | — | — |
| 41 | — | AClaude 4 Sonnet (Non-reasoning)闭源Anthropic | 32.7 | — | — | — |
| 42 | — | DDeepSeek V3.2 Exp (Reasoning)开源DeepSeek | 32.5 | — | — | — |
| 43 | — | XGrok 3 mini Reasoning (high)闭源xAI | 32.4 | — | — | — |
| 44 | — | AQwen3 Max Thinking闭源Alibaba | 32.3 | — | — | — |
| 45 | — | ZGLM-4.6 (Reasoning)开源Z AI | 32.2 | — | — | — |
| 46 | — | ANova 2.0 Pro Preview (low)闭源Amazon | 32.2 | — | — | — |
| 47 | — | LK-EXAONE (Reasoning)开源LG AI Research | 31.9 | — | — | — |
| 48 | — | DDeepSeek V3.2 (Non-reasoning)开源DeepSeek | 31.9 | — | — | — |
| 49 | — | AQwen3 Max闭源Alibaba | 31.0 | — | — | — |
| 50 | — | GGemini 2.5 Flash Preview (Sep '25) (Reasoning)闭源Google | 30.8 | — | — | — |
| 51 | — | AClaude 3.7 Sonnet (Non-reasoning)闭源Anthropic | 30.6 | — | — | — |
| 52 | — | AClaude 4.5 Haiku (Non-reasoning)闭源Anthropic | 30.5 | — | — | — |
| 53 | — | XMiMo-V2-Flash (Non-reasoning)开源Xiaomi | 30.1 | — | — | — |
| 54 | — | ANova 2.0 Lite (medium)闭源Amazon | 29.9 | — | — | — |
| 55 | — | ZGLM-4.6 (Non-reasoning)开源Z AI | 29.8 | — | — | — |
| 56 | — | AQwen3 235B A22B 2507 (Reasoning)开源Alibaba | 29.4 | — | — | — |
| 57 | — | DDeepSeek V3.2 Speciale开源DeepSeek | 29.0 | — | — | — |
| 58 | — | BERNIE 5.0 Thinking Preview闭源Baidu | 28.9 | — | — | — |
| 59 | — | KKimi K2 0905开源Kimi | 28.0 | — | — | — |
| 60 | — | DDeepSeek V3.2 Exp (Non-reasoning)开源DeepSeek | 28.0 | — | — | — |
| 61 | — | DDeepSeek V3.1 Terminus (Non-reasoning)开源DeepSeek | 28.0 | — | — | — |
| 62 | — | DDeepSeek V3.1 (Reasoning)开源DeepSeek | 27.9 | — | — | — |
| 63 | — | ANova 2.0 Omni (medium)闭源Amazon | 27.9 | — | — | — |
| 64 | — | SApriel-v1.6-15B-Thinker开源ServiceNow | 27.8 | — | — | — |
| 65 | — | DDeepSeek V3.1 (Non-reasoning)开源DeepSeek | 27.7 | — | — | — |
| 66 | — | AQwen3 VL 235B A22B (Reasoning)开源Alibaba | 27.5 | — | — | — |
| 67 | — | MMagistral Medium 1.2闭源Mistral | 27.5 | — | — | — |
| 68 | — | GGemini 2.5 Flash (Reasoning)闭源Google | 27.2 | — | — | — |
| 69 | — | OGPT-5.1 (Non-reasoning)闭源OpenAI | 27.2 | — | — | — |
| 70 | — | DDeepSeek R1 0528 (May '25)开源DeepSeek | 27.1 | — | — | — |
| 71 | — | ZGLM-4.5 (Reasoning)开源Z AI | 26.8 | — | — | — |
| 72 | — | OGPT-5 nano (high)闭源OpenAI | 26.7 | — | — | — |
| 73 | — | AQwen3 Next 80B A3B (Reasoning)开源Alibaba | 26.5 | — | — | — |
| 74 | — | OGPT-5 nano (medium)闭源OpenAI | 25.8 | — | — | — |
| 75 | — | OGPT-4.1闭源OpenAI | 25.8 | — | — | — |
| 76 | — | GGemini 2.5 Flash Preview (Sep '25) (Non-reasoning)闭源Google | 25.6 | — | — | — |
| 77 | — | XGrok 3闭源xAI | 25.2 | — | — | — |
| 78 | — | BSeed-OSS-36B-Instruct开源ByteDance Seed | 25.0 | — | — | — |
| 79 | — | ANova 2.0 Lite (low)闭源Amazon | 25.0 | — | — | — |
| 80 | — | AQwen3 Coder 480B A35B Instruct开源Alibaba | 24.8 | — | — | — |
| 81 | — | NNVIDIA Nemotron 3 Nano 30B A3B (Reasoning)开源NVIDIA | 24.7 | — | — | — |
| 82 | — | Ogpt-oss-20B (high)开源OpenAI | 24.7 | — | — | — |
| 83 | — | AQwen3 235B A22B 2507 Instruct开源Alibaba | 24.6 | — | — | — |
| 84 | — | OGPT-5 (minimal)闭源OpenAI | 24.5 | — | — | — |
| 85 | — | MMiniMax M1 80k开源MiniMax | 24.3 | — | — | — |
| 86 | — | ANova 2.0 Omni (low)闭源Amazon | 24.1 | — | — | — |
| 87 | — | NHyperCLOVA X SEED Think (32B)开源Naver | 24.0 | — | — | — |
| 88 | — | ZGLM-4.6V (Reasoning)开源Z AI | 23.8 | — | — | — |
| 89 | — | Ogpt-oss-120B (low)开源OpenAI | 23.8 | — | — | — |
| 90 | — | ZGLM-4.5-Air开源Z AI | 23.6 | — | — | — |
| 91 | — | XGrok 4.1 Fast (Non-reasoning)闭源xAI | 23.6 | — | — | — |
| 92 | — | ANova 2.0 Pro Preview (Non-reasoning)闭源Amazon | 23.5 | — | — | — |
| 93 | — | KMi:dm K 2.5 Pro开源Korea Telecom | 23.2 | — | — | — |
| 94 | — | LK-EXAONE (Non-reasoning)开源LG AI Research | 23.2 | — | — | — |
| 95 | — | OGPT-4.1 mini闭源OpenAI | 22.9 | — | — | — |
| 96 | — | AQwen3 30B A3B 2507 (Reasoning)开源Alibaba | 22.8 | — | — | — |
| 97 | — | XGrok 4 Fast (Non-reasoning)闭源xAI | 22.8 | — | — | — |
| 98 | — | DDeepSeek V3 0324开源DeepSeek | 22.6 | — | — | — |
| 99 | — | MMistral Large 3开源Mistral | 22.5 | — | — | — |
| 100 | — | GGemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)闭源Google | 22.3 | — | — | — |
| 101 | — | PINTELLECT-3开源Prime Intellect | 22.2 | — | — | — |
| 102 | — | MDevstral 2开源Mistral | 21.7 | — | — | — |
| 103 | — | OGPT-5 mini (minimal)闭源OpenAI | 21.5 | — | — | — |
| 104 | — | Ogpt-oss-20B (low)开源OpenAI | 21.2 | — | — | — |
| 105 | — | MMistral Medium 3.1闭源Mistral | 21.1 | — | — | — |
| 106 | — | MK2-V2 (high)开源MBZUAI Institute of Foundation Models | 21.1 | — | — | — |
| 107 | — | GGemini 2.5 Flash (Non-reasoning)闭源Google | 20.9 | — | — | — |
| 108 | — | AQwen3 VL 235B A22B Instruct开源Alibaba | 20.8 | — | — | — |
| 109 | — | AQwen3 Next 80B A3B Instruct开源Alibaba | 20.3 | — | — | — |
| 110 | — | AQwen3 Coder 30B A3B Instruct开源Alibaba | 20.2 | — | — | — |
| 111 | — | GGemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)闭源Google | 20.1 | — | — | — |
| 112 | — | AQwen3 235B A22B (Reasoning)开源Alibaba | 19.9 | — | — | — |
| 113 | — | ILing-1T开源InclusionAI | 19.7 | — | — | — |
| 113 | — | AQwen3 VL 30B A3B (Reasoning)开源Alibaba | 19.7 | — | — | — |
| 115 | — | NLlama Nemotron Super 49B v1.5 (Reasoning)开源NVIDIA | 19.4 | — | — | — |
| 116 | — | ANova Premier闭源Amazon | 19.2 | — | — | — |
| 117 | — | MK2-V2 (medium)开源MBZUAI Institute of Foundation Models | 19.2 | — | — | — |
| 118 | — | MDevstral Small 2开源Mistral | 19.1 | — | — | — |
| 119 | — | GGemini 2.0 Flash (Feb '25)闭源Google | 18.9 | — | — | — |
| 120 | — | MLlama 4 Maverick开源Meta | 18.9 | — | — | — |
| 121 | — | ANova 2.0 Lite (Non-reasoning)闭源Amazon | 18.8 | — | — | — |
| 122 | — | MDevstral Medium闭源Mistral | 18.8 | — | — | — |
| 123 | — | AQwen3 4B 2507 (Reasoning)开源Alibaba | 18.8 | — | — | — |
| 124 | — | DDeepSeek R1 (Jan '25)开源DeepSeek | 18.7 | — | — | — |
| 125 | — | GGemini 2.5 Flash-Lite (Reasoning)闭源Google | 17.9 | — | — | — |
| 126 | — | MLlama 3.1 Instruct 405B开源Meta | 17.5 | — | — | — |
| 127 | — | ZGLM-4.6V (Non-reasoning)开源Z AI | 17.4 | — | — | — |
| 128 | — | AQwen3 235B A22B (Non-reasoning)开源Alibaba | 17.3 | — | — | — |
| 129 | — | ANova 2.0 Omni (Non-reasoning)闭源Amazon | 17.2 | — | — | — |
| 130 | — | AQwen3 VL 32B Instruct开源Alibaba | 17.1 | — | — | — |
| 131 | — | AQwen3 32B (Reasoning)开源Alibaba | 17.1 | — | — | — |
| 132 | — | AQwen3 VL 8B (Reasoning)开源Alibaba | 17.0 | — | — | — |
| 133 | — | DDeepSeek V3 (Dec '24)开源DeepSeek | 17.0 | — | — | — |
| 134 | — | LEXAONE 4.0 32B (Reasoning)开源LG AI Research | 17.0 | — | — | — |
| 135 | — | AQwen3 14B (Reasoning)开源Alibaba | 16.8 | — | — | — |
| 136 | — | AQwen3 VL 30B A3B Instruct开源Alibaba | 16.3 | — | — | — |
| 137 | — | MMinistral 14B (Dec '25)开源Mistral | 16.3 | — | — | — |
| 138 | — | TFalcon-H1R-7B开源TII UAE | 16.2 | — | — | — |
| 139 | — | AQwen3 30B A3B (Reasoning)开源Alibaba | 15.8 | — | — | — |
| 140 | — | AQwen3 Omni 30B A3B (Reasoning)开源Alibaba | 15.8 | — | — | — |
| 141 | — | MDevstral Small (Jul '25)开源Mistral | 15.7 | — | — | — |
| 142 | — | USolar Pro 2 (Reasoning)闭源Upstage | 15.5 | — | — | — |
| 143 | — | NNVIDIA Nemotron Nano 9B V2 (Reasoning)开源NVIDIA | 15.5 | — | — | — |
| 144 | — | NLlama Nemotron Super 49B v1.5 (Non-reasoning)开源NVIDIA | 15.4 | — | — | — |
| 144 | — | AQwen3 30B A3B 2507 Instruct开源Alibaba | 15.4 | — | — | — |
| 146 | — | MMistral Small 3.2开源Mistral | 15.3 | — | — | — |
| 147 | — | MK2-V2 (low)开源MBZUAI Institute of Foundation Models | 15.3 | — | — | — |
| 148 | — | NNVIDIA Nemotron Nano 12B v2 VL (Reasoning)开源NVIDIA | 15.1 | — | — | — |
| 149 | — | MMinistral 8B (Dec '25)开源Mistral | 15.1 | — | — | — |
| 150 | — | MLlama 3.3 Instruct 70B开源Meta | 15.1 | — | — | — |
| 151 | — | AQwen3 VL 8B Instruct开源Alibaba | 14.6 | — | — | — |
| 152 | — | AOlmo 3.1 32B Think开源Allen Institute for AI | 14.2 | — | — | — |
| 152 | — | USolar Pro 2 (Non-reasoning)闭源Upstage | 14.2 | — | — | — |
| 154 | — | MLlama 4 Scout开源Meta | 14.0 | — | — | — |
| 155 | — | NLlama 3.1 Nemotron Instruct 70B开源NVIDIA | 14.0 | — | — | — |
| 156 | — | OGPT-5 nano (minimal)闭源OpenAI | 14.0 | — | — | — |
| 157 | — | NNVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)开源NVIDIA | 13.9 | — | — | — |
| 158 | — | NNVIDIA Nemotron Nano 9B V2 (Non-reasoning)开源NVIDIA | 13.8 | — | — | — |
| 159 | — | CCommand A开源Cohere | 13.7 | — | — | — |
| 160 | — | OGPT-4.1 nano闭源OpenAI | 13.4 | — | — | — |
| 161 | — | AQwen3 14B (Non-reasoning)开源Alibaba | 13.4 | — | — | — |
| 162 | — | ZGLM-4.5V (Non-reasoning)开源Z AI | 13.2 | — | — | — |
| 163 | — | AQwen3 4B 2507 Instruct开源Alibaba | 13.1 | — | — | — |
| 164 | — | GGemini 2.5 Flash-Lite (Non-reasoning)闭源Google | 13.0 | — | — | — |
| 165 | — | ANova Lite闭源Amazon | 12.5 | — | — | — |
| 166 | — | AQwen3 30B A3B (Non-reasoning)开源Alibaba | 12.4 | — | — | — |
| 167 | — | MLlama 3.1 Instruct 8B开源Meta | 12.2 | — | — | — |
| 168 | — | MMinistral 3B (Dec '25)开源Mistral | 12.1 | — | — | — |
| 169 | — | LEXAONE 4.0 32B (Non-reasoning)开源LG AI Research | 12.0 | — | — | — |
| 170 | — | IGranite 4.0 H Small开源IBM | 11.5 | — | — | — |
| 171 | — | AQwen3 8B (Non-reasoning)开源Alibaba | 10.8 | — | — | — |
| 172 | — | AQwen3 Omni 30B A3B Instruct开源Alibaba | 10.7 | — | — | — |
| 173 | — | GGemma 3 27B Instruct开源Google | 10.5 | — | — | — |
| 174 | — | NNVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)开源NVIDIA | 10.4 | — | — | — |
| 175 | — | AJamba 1.7 Large开源AI21 Labs | 9.7 | — | — | — |
| 176 | — | GGemma 3 12B Instruct开源Google | 9.4 | — | — | — |
| 177 | — | LExaone 4.0 1.2B (Reasoning)开源LG AI Research | 8.6 | — | — | — |
| 178 | — | LExaone 4.0 1.2B (Non-reasoning)开源LG AI Research | 8.6 | — | — | — |
| 179 | — | IGranite 4.0 H 1B开源IBM | 8.4 | — | — | — |
| 180 | — | AOlmo 3 7B Instruct开源Allen Institute for AI | 8.2 | — | — | — |
| 181 | — | AJamba 1.7 Mini开源AI21 Labs | 7.9 | — | — | — |
| 182 | — | LLFM2 2.6B开源Liquid AI | 7.9 | — | — | — |
| 183 | — | IGranite 4.0 1B开源IBM | 7.7 | — | — | — |
| 184 | — | LLFM2 8B A1B开源Liquid AI | 7.3 | — | — | — |
| 185 | — | IGranite 4.0 350M开源IBM | 7.2 | — | — | — |
| 186 | — | LLFM2 1.2B开源Liquid AI | 6.8 | — | — | — |
| 187 | — | AQwen3 1.7B (Non-reasoning)开源Alibaba | 6.8 | — | — | — |
| 188 | — | GGemma 3 4B Instruct开源Google | 6.6 | — | — | — |
| 189 | — | GGemma 3n E4B Instruct开源Google | 6.3 | — | — | — |
| 190 | — | AQwen3 0.6B (Non-reasoning)开源Alibaba | 6.1 | — | — | — |
| 191 | — | IGranite 4.0 H 350M开源IBM | 6.0 | — | — | — |