Design Arena
Design Arena 设计代码排行榜由 Arcada Labs 提供，汇总 Website、UI Component、Game Dev、Data Visualization 等代码子类别的匿名投票，并通过 Bradley-Terry 模型计算综合排名。
榜首模型
Claude Opus 5.5
最高得分
1395.00
模型数量
171
数据版本
2026年10月01日
数据来源: Arcada Labs
排名总表
| 排名 | 模型名称 | 得分 | 95% CI | 投票数 | 机构 | 许可证 |
|---|---|---|---|---|---|---|
 Claude Opus 5.5Anthropic | 1395.00 | +/-20 | 1,298 | Anthropic | Proprietary | |
 GPT-6 Astra (xhigh)OpenAI | 1389.00 | +/-9 | 6,610 | OpenAI | Proprietary | |
 Kimi K3Moonshot AI | 1372.00 | +/-8.3 | 7,620 | Moonshot AI | Open Source | |
| 4 | GPT-6 Astra (max)OpenAI | 1362.00 | +/-18.1 | 1,523 | OpenAI | Proprietary |
| 5 |  Muse Spark 1.3 (xhigh)Meta | 1359.00 | +/-4.8 | 26,469 | Meta | Proprietary |
| 6 | Muse Spark 1.3 MaxMeta | 1358.00 | +/-5.2 | 21,850 | Meta | Proprietary |
| 7 | GPT-5.6 Sol (xhigh)OpenAI | 1346.00 | +/-7.3 | 9,739 | OpenAI | Proprietary |
| 8 | Claude Fable 5.1Anthropic | 1342.00 | +/-7.5 | 9,335 | Anthropic | Proprietary |
| 9 | Claude Opus 5Anthropic | 1329.00 | +/-5.8 | 15,990 | Anthropic | Proprietary |
| 10 |  DeepSeek-V4.1-FlashDeepSeek | 1329.00 | +/-5.1 | 23,191 | DeepSeek | Open Source |
| 11 | GPT-5.6 Sol (medium)OpenAI | 1325.00 | +/-4.8 | 24,115 | OpenAI | Proprietary |
| 12 | MiMo-V2.6-ProXiaomi | 1325.00 | +/-5.4 | 18,979 | Xiaomi | Open Source |
| 13 |  GLM-5.3智谱AI | 1318.00 | +/-5.8 | 16,228 | 智谱AI | Open Source |
| 14 | Muse Spark 1.2Facebook AI研究实验室 | 1318.00 | +/-4.4 | 32,086 | Facebook AI研究实验室 | Proprietary |
| 15 | Claude Fable 5Anthropic | 1317.00 | +/-5.2 | 20,186 | Anthropic | Proprietary |
| 16 |  Gemini 3.7 FlashGoogle DeepMind | 1315.00 | +/-3.5 | 58,599 | Google DeepMind | Proprietary |
| 17 | Gemini 3.8 FlashGoogle | 1311.00 | +/-4.7 | 27,867 | Proprietary | |
| 18 |  Qwen3.8 MaxAlibaba | 1300.00 | +/-7.5 | 8,888 | Alibaba | Open Source |
| 19 |  Grok 4.6xAI | 1299.00 | +/-4.2 | 33,210 | xAI | Proprietary |
| 20 | GLM-5.2智谱AI | 1297.00 | +/-3.8 | 43,349 | 智谱AI | Open Source |
| 21 | Gemini 3.6 FlashGoogle | 1296.00 | +/-4.4 | 29,244 | Proprietary | |
| 22 | Claude Opus 4.6Anthropic | 1295.00 | +/-4.5 | 27,401 | Anthropic | Proprietary |
| 23 | Claude Opus 4.6 (thinking)Anthropic | 1292.00 | +/-4.8 | 23,808 | Anthropic | Proprietary |
| 24 | GPT-6 SolOpenAI | 1292.00 | +/-15.3 | 2,047 | OpenAI | Proprietary |
| 25 | GLM-5.3-FlashZhipu AI | 1289.00 | +/-4.7 | 26,357 | Zhipu AI | Open Source |
| 26 | Deca 2.6 UltraGenLabs | 1284.00 | +/-14 | 2,447 | GenLabs | Proprietary |
| 27 | Claude Sonnet 5Anthropic | 1283.00 | +/-5 | 21,742 | Anthropic | Proprietary |
| 28 | Muse Spark 1.1Facebook AI研究实验室 | 1283.00 | +/-5.1 | 20,655 | Facebook AI研究实验室 | Proprietary |
| 29 | Claude Sonnet 4.6Anthropic | 1282.00 | +/-4.6 | 26,314 | Anthropic | Proprietary |
| 30 | Qwen3.7 Max阿里巴巴 | 1282.00 | +/-4.1 | 33,404 | 阿里巴巴 | Proprietary |
| 31 | Grok 4.5xAI | 1282.00 | +/-4.2 | 34,186 | xAI | Proprietary |
| 32 | GPT-6 LunaOpenAI | 1281.00 | +/-12 | 3,397 | OpenAI | Proprietary |
| 33 | Deca 2.5 MiniGenLabs | 1278.00 | +/-10.8 | 4,205 | GenLabs | Proprietary |
| 34 | Kimi K2.6Moonshot AI | 1278.00 | +/-4.1 | 34,037 | Moonshot AI | Open Source |
| 35 | GLM 5.1智谱AI | 1274.00 | +/-4.5 | 27,184 | 智谱AI | Open Source |
| 36 | MiMo-V2.5-ProXiaomi | 1274.00 | +/-4.6 | 26,263 | Xiaomi | Open Source |
| 37 | Qwen3.7-Plus阿里巴巴 | 1274.00 | +/-5.7 | 16,127 | 阿里巴巴 | Proprietary |
| 38 | Gemini 3.5 FlashGoogle DeepMind | 1271.00 | +/-4.1 | 33,933 | Google DeepMind | Proprietary |
| 39 | Deca 2.5 UltraGenLabs | 1269.00 | +/-11.1 | 3,939 | GenLabs | Proprietary |
| 40 | GLM-5-Turbo智谱AI | 1269.00 | +/-3.8 | 39,942 | 智谱AI | Proprietary |
| 41 | GPT-5.5OpenAI | 1266.00 | +/-4.3 | 30,546 | OpenAI | Proprietary |
| 42 | MiMo-V2.5Xiaomi | 1266.00 | +/-3.7 | 42,812 | Xiaomi | Open Source |
| 43 | Kimi K2.7 CodeMoonshot AI | 1260.00 | +/-5.1 | 20,786 | Moonshot AI | Open Source |
| 44 | Claude Opus 4.8Anthropic | 1259.00 | +/-4.3 | 30,046 | Anthropic | Proprietary |
| 45 | Muse SparkFacebook AI研究实验室 | 1258.00 | +/-10.9 | 4,250 | Facebook AI研究实验室 | Proprietary |
| 46 |  MiniMax M3MiniMaxAI | 1254.00 | +/-4.5 | 27,124 | MiniMaxAI | Open Source |
| 47 | GLM-5智谱AI | 1252.00 | +/-3.6 | 45,362 | 智谱AI | Open Source |
| 48 | Gemini 3.0 Pro (Preview 11-2025)Google DeepMind | 1250.00 | +/-4.8 | 25,342 | Google DeepMind | Proprietary |
| 49 | Opus 4.5Anthropic | 1248.00 | +/-4 | 35,206 | Anthropic | Proprietary |
| 50 | GPT-5.6 TerraOpenAI | 1247.00 | +/-3.2 | 71,396 | OpenAI | Proprietary |
| 51 | DeepSeek-V4-ProDeepSeek-AI | 1246.00 | +/-4.5 | 27,256 | DeepSeek-AI | Open Source |
| 52 | Kimi K2.5 (thinking)Moonshot AI | 1242.00 | +/-3.7 | 41,629 | Moonshot AI | Open Source |
| 53 | Qwen 3.6 Plus Preview阿里巴巴 | 1242.00 | +/-4 | 36,716 | 阿里巴巴 | Proprietary |
| 54 | MiniMax-M2.7MiniMaxAI | 1240.00 | +/-3.7 | 44,124 | MiniMaxAI | Open Source |
| 55 | GPT-5.6 LunaOpenAI | 1239.00 | +/-5.4 | 18,302 | OpenAI | Proprietary |
| 56 | Nex N2 ProNex AGI | 1238.00 | +/-6.3 | 13,047 | Nex AGI | Open Source |
| 57 | Gemini 3.1 Pro PreviewGoogle DeepMind | 1237.00 | +/-3.5 | 47,230 | Google DeepMind | Proprietary |
| 58 | DeepSeek-V4-FlashDeepSeek-AI | 1236.00 | +/-7 | 10,423 | DeepSeek-AI | Open Source |
| 59 | GLM-5V-Turbo智谱AI | 1235.00 | +/-3.7 | 44,624 | 智谱AI | Proprietary |
| 60 |  Hy3腾讯AI实验室 | 1234.00 | +/-4.6 | 26,324 | 腾讯AI实验室 | Open Source |
| 61 | Grok 4.20 Beta (Reasoning)xAI | 1227.00 | +/-3.9 | 37,038 | xAI | Proprietary |
| 62 | GLM-4.7智谱AI | 1223.00 | +/-3.5 | 46,843 | 智谱AI | Open Source |
| 63 | GPT-5.4 (Design Skill, Medium)OpenAI | 1218.00 | +/-7.2 | 9,610 | OpenAI | Proprietary |
| 64 | GPT-5.4 (medium)OpenAI | 1216.00 | +/-5.3 | 18,950 | OpenAI | Proprietary |
| 65 | InklingThinking Machines Lab | 1214.00 | +/-12.4 | 3,187 | Thinking Machines Lab | Open Source |
| 66 | MiniMax M2.5MiniMaxAI | 1213.00 | +/-6.7 | 11,506 | MiniMaxAI | Open Source |
| 67 | Grok 4.20 BetaxAI | 1212.00 | +/-3.8 | 40,209 | xAI | Proprietary |
| 68 | DeepSeek-V4-FlashDeepSeek-AI | 1211.00 | +/-4.4 | 28,093 | DeepSeek-AI | Open Source |
| 69 | InklingThinking Machines Lab | 1210.00 | +/-6.4 | 12,938 | Thinking Machines Lab | Open Source |
| 70 | M2.1MiniMaxAI | 1196.00 | +/-5 | 20,805 | MiniMaxAI | Open Source |
| 71 | Gemini 3.0 FlashGoogle DeepMind | 1195.00 | +/-10.6 | 4,414 | Google DeepMind | Proprietary |
| 72 | Grok 4.3 BetaxAI | 1193.00 | +/-3.9 | 37,879 | xAI | Proprietary |
| 73 | Claude Sonnet 4.5Anthropic | 1189.00 | +/-3.6 | 45,435 | Anthropic | Proprietary |
| 74 | Solar Pro 4Upstage | 1189.00 | +/-5.6 | 17,856 | Upstage | Proprietary |
| 75 | Claude Sonnet 4.5 (thinking)Anthropic | 1188.00 | +/-3.6 | 44,192 | Anthropic | Proprietary |
| 76 |  Step 3.7 FlashStepFunAI | 1187.00 | +/-4.5 | 27,413 | StepFunAI | Open Source |
| 77 | GPT-5.4 (low)OpenAI | 1186.00 | +/-5 | 21,146 | OpenAI | Proprietary |
| 78 | Qwen3.5-397B-A17B阿里巴巴 | 1186.00 | +/-7.8 | 8,130 | 阿里巴巴 | Open Source |
| 79 | GLM-4.7-Flash智谱AI | 1184.00 | +/-6.6 | 11,706 | 智谱AI | Open Source |
| 80 | GPT-5.4OpenAI | 1184.00 | +/-4.8 | 23,605 | OpenAI | Proprietary |
| 81 | Hy3 (low)腾讯AI实验室 | 1183.00 | +/-11.1 | 4,070 | 腾讯AI实验室 | Open Source |
| 82 | Claude Sonnet 3.7Anthropic | 1182.00 | +/-5.8 | 15,259 | Anthropic | Proprietary |
| 83 | DeepSeek-V3.1 (thinking)DeepSeek-AI | 1181.00 | +/-5.7 | 16,257 | DeepSeek-AI | Open Source |
| 84 | Opus 4.1 (thinking)Anthropic | 1177.00 | +/-5.7 | 15,715 | Anthropic | Proprietary |
| 85 | DeepSeek V3.2-ExpDeepSeek-AI | 1177.00 | +/-5.2 | 19,487 | DeepSeek-AI | Open Source |
| 86 | GPT-5.1 (high)OpenAI | 1177.00 | +/-5.6 | 16,116 | OpenAI | Proprietary |
| 87 | GPT-5.2 (medium)OpenAI | 1176.00 | +/-4.4 | 27,935 | OpenAI | Proprietary |
| 88 | GPT-5.2 (None)OpenAI | 1175.00 | +/-4.3 | 28,878 | OpenAI | Proprietary |
| 89 | GPT-5 (high)OpenAI | 1174.00 | +/-6.2 | 13,485 | OpenAI | Proprietary |
| 90 | Qwen3.5 Plus (0215)阿里巴巴 | 1173.00 | +/-5.3 | 18,500 | 阿里巴巴 | Proprietary |
| 91 | GLM-4.6智谱AI | 1172.00 | +/-5.5 | 17,086 | 智谱AI | Open Source |
| 92 | GPT-5.2 (low)OpenAI | 1172.00 | +/-4.6 | 25,458 | OpenAI | Proprietary |
| 93 | Opus 4.1Anthropic | 1171.00 | +/-3.8 | 39,744 | Anthropic | Proprietary |
| 94 | GLM-4.5智谱AI | 1171.00 | +/-5.2 | 19,653 | 智谱AI | Open Source |
| 95 | GPT-5 (minimal)OpenAI | 1171.00 | +/-4.1 | 32,967 | OpenAI | Proprietary |
| 96 | DeepSeek V3.2DeepSeek-AI | 1170.00 | +/-4.3 | 29,125 | DeepSeek-AI | Open Source |
| 97 | Claude Opus 4Anthropic | 1167.00 | +/-5.6 | 16,668 | Anthropic | Proprietary |
| 98 | GPT-5.1 (medium)OpenAI | 1167.00 | +/-5 | 21,347 | OpenAI | Proprietary |
| 99 | GPT-5.1 (low)OpenAI | 1162.00 | +/-4.9 | 22,192 | OpenAI | Proprietary |
| 100 | MiMo-V2-FlashXiaomi | 1162.00 | +/-4 | 35,226 | Xiaomi | Open Source |
| 101 | Gemini 2.5 ProGoogle DeepMind | 1158.00 | +/-8.5 | 7,220 | Google DeepMind | Proprietary |
| 102 | GPT-5.1 CodexOpenAI | 1158.00 | +/-15.6 | 2,005 | OpenAI | Proprietary |
| 103 | GPT-5.1 (None)OpenAI | 1157.00 | +/-4.9 | 22,295 | OpenAI | Proprietary |
| 104 | GPT-5.2 (high)OpenAI | 1156.00 | +/-10.8 | 4,150 | OpenAI | Proprietary |
| 105 | GPT-5.3 CodexOpenAI | 1151.00 | +/-5.7 | 15,675 | OpenAI | Proprietary |
| 106 |  Nemotron 3 UltraNVIDIA | 1150.00 | +/-5 | 23,370 | NVIDIA | Open Source |
| 107 | v0-1.5-lgVercel | 1150.00 | +/-21.4 | 1,192 | Vercel | Proprietary |
| 108 | Qwen3-Coder-480B-A35B阿里巴巴 | 1149.00 | +/-16.3 | 1,955 | 阿里巴巴 | Open Source |
| 109 |  Mistral Large 3MistralAI | 1148.00 | +/-4.2 | 30,429 | MistralAI | Open Source |
| 110 | Claude Sonnet 4Anthropic | 1147.00 | +/-5.4 | 17,542 | Anthropic | Proprietary |
| 111 | DeepSeek-R1-0528DeepSeek-AI | 1145.00 | +/-5.4 | 17,951 | DeepSeek-AI | Open Source |
| 112 | AGI-01 SwiftLucidQuery | 1143.00 | +/-20.2 | 1,249 | LucidQuery | Proprietary |
| 113 | GLM-4.5-Air智谱AI | 1143.00 | +/-5.5 | 17,253 | 智谱AI | Open Source |
| 114 | Claude Sonnet 4 (thinking)Anthropic | 1142.00 | +/-5.6 | 16,246 | Anthropic | Proprietary |
| 115 | MiniMax M2MiniMaxAI | 1141.00 | +/-6.8 | 10,828 | MiniMaxAI | Open Source |
| 116 | AesCoder-4BDesignFlow | 1131.00 | +/-3.8 | 39,834 | DesignFlow | Open Source |
| 117 | Mistral Medium 3.5MistralAI | 1127.00 | +/-6.9 | 10,894 | MistralAI | Open Source |
| 118 | Mistral Medium 3.1 (2508)Mistral | 1126.00 | +/-4.4 | 27,827 | Mistral | Proprietary |
| 119 | Trinity Large ThinkingArcee AI | 1123.00 | +/-6.4 | 12,837 | Arcee AI | Open Source |
| 120 | Haiku 4.5Anthropic | 1121.00 | +/-4 | 35,541 | Anthropic | Proprietary |
| 121 | GPT-5-miniOpenAI | 1121.00 | +/-3.9 | 37,089 | OpenAI | Proprietary |
| 122 | DeepSeek-V3.1DeepSeek-AI | 1118.00 | +/-5.1 | 20,334 | DeepSeek-AI | Open Source |
| 123 | Qwen3-Max-Thinking阿里巴巴 | 1116.00 | +/-4.1 | 33,368 | 阿里巴巴 | Proprietary |
| 124 | DeepSeek-V3-0324DeepSeek-AI | 1114.00 | +/-5.2 | 19,271 | DeepSeek-AI | Open Source |
| 125 | Prime Intellect: INTELLECT-3Prime Intellect | 1113.00 | +/-4.2 | 31,349 | Prime Intellect | Open Source |
| 126 | Gemini 2.5 Flash-Preview-09-2025Google DeepMind | 1110.00 | +/-5.2 | 19,295 | Google DeepMind | Proprietary |
| 127 | Grok 4 FastxAI | 1106.00 | +/-3.9 | 36,646 | xAI | Proprietary |
| 128 | Kimi K2 0905Moonshot AI | 1104.00 | +/-17.9 | 1,504 | Moonshot AI | Open Source |
| 129 | GPT-5.1 Codex MiniOpenAI | 1100.00 | +/-4.1 | 33,830 | OpenAI | Proprietary |
| 130 | Grok 4.1 FastxAI | 1098.00 | +/-4.1 | 33,602 | xAI | Proprietary |
| 131 | Grok 4.1 Fast (reasoning)xAI | 1093.00 | +/-4.3 | 31,133 | xAI | Proprietary |
| 132 | GPT-5-NanoOpenAI | 1090.00 | +/-8.6 | 6,703 | OpenAI | Proprietary |
| 133 | Kimi K2 Turbo PreviewMoonshot AI | 1089.00 | +/-15.2 | 2,094 | Moonshot AI | Open Source |
| 134 | Gemini 2.5 Flash-Lite-Preview-09-2025Google DeepMind | 1087.00 | +/-8.5 | 6,850 | Google DeepMind | Proprietary |
| 135 | Gemini 3.1 Flash-LiteGoogle DeepMind | 1078.00 | +/-4.9 | 23,537 | Google DeepMind | Proprietary |
| 136 |  Phi-3-medium 14B-previewMicrosoft Azure | 1075.00 | +/-8.9 | 6,390 | Microsoft Azure | Proprietary |
| 137 | Ministral 3 14BMistralAI | 1071.00 | +/-14.4 | 2,380 | MistralAI | Open Source |
| 138 | Gemini 2.5 FlashGoogle DeepMind | 1065.00 | +/-8.5 | 6,971 | Google DeepMind | Proprietary |
| 139 | v0-1.5-mdVercel | 1063.00 | +/-6.8 | 11,081 | Vercel | Proprietary |
| 140 | Grok 3xAI | 1059.00 | +/-4.6 | 26,865 | xAI | Proprietary |
| 141 | Ministral 3 8BMistralAI | 1059.00 | +/-14.3 | 2,427 | MistralAI | Open Source |
| 142 | Grok 4 Fast (reasoning)xAI | 1056.00 | +/-4 | 37,370 | xAI | Proprietary |
| 143 | Qwen3-235B-A22B-2507阿里巴巴 | 1044.00 | +/-8.6 | 6,939 | 阿里巴巴 | Open Source |
| 144 | Kimi K2Moonshot AI | 1039.00 | +/-19.5 | 1,352 | Moonshot AI | Open Source |
| 145 | Magistral Medium 1.2 (2509)Mistral | 1039.00 | +/-9.4 | 5,854 | Mistral | Proprietary |
| 146 | Qwen3-235B-A22B-Thinking-2507Alibaba | 1039.00 | +/-9.1 | 6,175 | Alibaba | Open Source |
| 147 | GPT-4.1OpenAI | 1032.00 | +/-17.3 | 1,747 | OpenAI | Proprietary |
| 148 | OpenAI o3OpenAI | 1026.00 | +/-19.5 | 1,365 | OpenAI | Proprietary |
| 149 | Mercury 2Inception | 1023.00 | +/-5.8 | 22,317 | Inception | Proprietary |
| 150 | Grok 4xAI | 1022.00 | +/-4.9 | 24,016 | xAI | Proprietary |
| 151 | Devstral MediumMistralAI | 1018.00 | +/-8.6 | 7,153 | MistralAI | Proprietary |
| 152 | Ministral 3 3B (2512)Mistral | 1016.00 | +/-13.5 | 2,852 | Mistral | Open Source |
| 153 | Codestral 2508Mistral | 1013.00 | +/-8.8 | 6,734 | Mistral | Proprietary |
| 154 | Qwen3-235B-A22B阿里巴巴 | 1008.00 | +/-10.1 | 5,158 | 阿里巴巴 | Open Source |
| 155 | Grok Code Fast 1xAI | 1005.00 | +/-11.1 | 4,299 | xAI | Proprietary |
| 156 | GPT-4.1 miniOpenAI | 1000.00 | +/-18.3 | 1,566 | OpenAI | Proprietary |
| 157 | Magistral Small 1.2 (2509)Mistral | 992.00 | +/-9.2 | 6,447 | Mistral | Open Source |
| 158 | OpenAI o4 - miniOpenAI | 982.00 | +/-16.2 | 2,011 | OpenAI | Proprietary |
| 159 | Olmo 3.1 32B ThinkAllen AI | 981.00 | +/-6.3 | 16,164 | Allen AI | Open Source |
| 160 | GPT-4.1 nanoOpenAI | 969.00 | +/-16.9 | 1,901 | OpenAI | Proprietary |
| 161 | GPT OSS 120BOpenAI | 969.00 | +/-10.3 | 5,272 | OpenAI | Open Source |
| 162 | Qwen3-30B-A3B阿里巴巴 | 948.00 | +/-14.6 | 2,568 | 阿里巴巴 | Open Source |
| 163 | Grok 3 minixAI | 936.00 | +/-8.8 | 7,621 | xAI | Proprietary |
| 164 | Llama 3.1 Nemotron Ultra 253BNVIDIA | 935.00 | +/-13.8 | 3,173 | NVIDIA | Open Source |
| 165 | Mistral-Small-3.2MistralAI | 913.00 | +/-20.8 | 1,243 | MistralAI | Open Source |
| 166 | Llama 4 MaverickFacebook AI研究实验室 | 886.00 | +/-18.4 | 1,678 | Facebook AI研究实验室 | Open Source |
| 167 | Mistral Large 2.1 (2411)Mistral | 869.00 | +/-21 | 1,317 | Mistral | Proprietary |
| 168 | GPT-4oOpenAI | 867.00 | +/-18.2 | 1,780 | OpenAI | Proprietary |
| 169 | Codestral 2 (2501)Mistral | 840.00 | +/-20.7 | 1,444 | Mistral | Open Source |
| 170 | Devstral Small 1.1MistralAI | 813.00 | +/-22.6 | 1,250 | MistralAI | Open Source |
| 171 | Llama 4 ScoutFacebook AI研究实验室 | 796.00 | +/-22.7 | 1,275 | Facebook AI研究实验室 | Open Source |
数据仅供参考，以官方来源为准。模型名称旁的链接可跳转到 DataLearner 模型详情页。
关于本榜单
本榜单数据来源于Design Arena，由 Y Combinator 支持的 Arcada Labs 开发，是专注于评测 AI 设计代码生成能力的众包匿名对战平台。
与 LMArena 评测通用文本和编程能力不同，Design Arena 的代码榜专门考察模型生成具有视觉呈现效果的前端代码的能力。平台将代码任务细分为 Website、UI 组件、游戏开发、数据可视化、SVG、Web App、移动端等多个子类别，每个子类别均有独立排行。
本页展示的是 Code Categories 综合榜，即将所有子类别的用户投票混合汇总后，统一用 Bradley-Terry 模型（类 Elo 算法）计算出的综合排名。每票等权，不对各子类别做加权处理，因此投票量较大的子类别（如 Website）对综合分数的影响更大。得分越高，代表模型在设计代码生成场景下的综合人类偏好越强。
常见问题 (FAQ)
01
什么是 Design Arena？
Design Arena 是 Arcada Labs 开发的设计代码生成匿名评测平台。本榜单汇总 Website、UI 组件、游戏开发、数据可视化等代码子类别的投票。
02
Design Arena 与 LMArena Coding Arena 有什么区别？
LMArena Coding Arena 主要评测通用编程能力，例如代码生成、调试和算法实现；Design Arena 专注于具有视觉呈现效果的前端设计代码，例如 HTML 页面、交互 UI、图表、SVG 和原型。
03
排名方法论是什么？
Arcada Labs 将各代码子类别的原始投票混合后运行 Bradley-Terry 模型。每票等权，不按子类别单独加权，因此投票量较大的子类别会对综合分数产生更大影响。
04
哪类模型在设计代码场景表现更好？
具备强视觉理解和前端代码生成能力的大模型通常表现更好。针对 UI 和代码生成优化的专项模型，在布局、交互和视觉细节任务上也可能有突出表现。