Best LLM for web design: Design Arena: websites ranking
- Muse Wonder leads
- 1360
- models measured
- 131
- lead over #2
- +4 Elo
- median
- 1199
Leader on Design Arena: websites · read October 20
As of October 20, 2026, Meta Muse Wonder leads web design with 1360 on Design Arena: websites, ahead of Meta Muse Spark 1.3 (1356) and OpenAI GPT-6 Astra (1338). This ranking is one public board, shown as published, and is not part of the Vibe Coding Index.
| Rank | Model | Design Arena: websites | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|
| 1 | Muse Wonder Meta | 1360 | 64.5 | $20.00 | |
| 2 | Muse Spark 1.3 Meta | 1356 | 79.7 | $4.25 | |
| 3 | GPT-6 Astra OpenAI | 1338 | 88.7 | $50.00 | |
| 4 | Claude Opus 5.5 Anthropic | 1336 | 92.2 | $20.00 | |
| 5 | Kimi K3 Moonshot AI | 1335 | 61.8 | $11.936 | |
| 6 | Claude Sonnet 5.5 Anthropic | 1326 | 88.9 | $10.00 | |
| 7 | GPT-5.6 Sol OpenAI | 1321 | 76.7 | $20.00 | |
| 8 | GPT-6.1 Sol OpenAI | 1320 | 88.7 | $10.00 | |
| 9 | Gemini 4 Argon Google | 1315 | 80.5 | $10.00 | |
| 10 | Claude Fable 5.1 Anthropic | 1313 | 89.1 | $50.00 | |
| 11 | Muse Spark 1.2 Meta | 1313 | 65.1 | $4.25 | |
| 12 | DeepSeek V4.1 Flash 1017 DeepSeek | 1312 | 83.3 | $0.60 | |
| 13 | Claude Opus 5 Anthropic | 1310 | 81.8 | $25.00 | |
| 14 | DeepSeek V4.1 Flash DeepSeek | 1309 | 77.2 | $0.0112 | |
| 15 | MiMo-V2.6-Pro Xiaomi | 1308 | 72.6 | $0.87 | |
| 16 | Muse Spark Open Meta | 1306 | 62.1 | $4.25 | |
| 17 | Gemini 3.7 Flash Google | 1304 | 69.9 | $3.75 | |
| 18 | Gemini 3.8 Flash Google | 1301 | 70.5 | $4.125 | |
| 19 | Gemini 3.6 Flash Google | 1300 | 48.3 | $3.75 | |
| 20 | Claude Fable 5 Anthropic | 1296 | 82.1 | $50.00 | |
| 21 | GLM 5.3 Z.ai | 1295 | 73.1 | $1.40 | |
| 22 | GPT-6 Sol OpenAI | 1292 | 82.8 | $10.00 | |
| 23 | Claude Opus 4.6 Anthropic | 1291 | 66.9 | $25.00 | |
| 24 | GLM 5.1 Z.ai | 1290 | 44.0 | $3.036 | |
| 25 | Grok 4.6 xAI | 1289 | 76.4 | $6.00 | |
| 26 | Claude Sonnet 4.6 Anthropic | 1285 | 57.8 | $15.00 | |
| 27 | GLM 5.2 Z.ai | 1285 | 57.7 | $2.42 | |
| 28 | Grok 4.5 xAI | 1281 | 75.5 | $6.00 | |
| 29 | Qwen3.8 Max (0902) Qwen | 1276 | 63.5 | $6.00 | |
| 30 | Claude Sonnet 5 Anthropic | 1274 | 63.5 | $10.00 | |
| 31 | Qwen3.7 Max Qwen | 1274 | 45.1 | $4.425 | |
| 32 | GLM 5.3 Flash Z.ai | 1272 | 71.0 | $0.16 | |
| 33 | GPT-6 Luna OpenAI | 1271 | 71.4 | $0.50 | |
| 34 | Muse Spark 1.1 Meta | 1270 | 48.9 | $4.25 | |
| 35 | Qwen3.7 Plus Qwen | 1270 | Not benchmarked | $1.28 | |
| 36 | Kimi K2.6 Moonshot AI | 1269 | 44.2 | $3.41 | |
| 37 | GPT-5.6 Terra OpenAI | 1269 | 69.0 | $12.00 | |
| 38 | Gemini 3.5 Flash Google | 1268 | 42.8 | $9.00 | |
| 39 | MiMo-V2.5-Pro Xiaomi | 1267 | 40.0 | $0.87 | |
| 40 | MiMo-V2.5 Xiaomi | 1266 | 34.0 | $0.28 | |
| 41 | Kimi K2.7 Code Moonshot AI | 1264 | 39.1 | $3.35 | |
| 42 | Claude Haiku 5.5 Anthropic | 1263 | 75.9 | $0.50 | |
| 43 | GLM 5 Turbo Z.ai | 1262 | Not benchmarked | $4.00 | |
| 44 | Claude Opus 4.8 Anthropic | 1259 | 66.0 | $25.00 | |
| 45 | MiniMax M3 MiniMax | 1258 | 57.8 | $1.20 | |
| 46 | Mistral Large 4 Mistral AI | 1258 | 55.3 | $2.09 | |
| 47 | GPT-5.5 OpenAI | 1257 | 67.8 | $30.00 | |
| 48 | GPT-5.6 Luna OpenAI | 1256 | 61.5 | $1.20 | |
| 49 | Grok 4.7 xAI | 1253 | 75.0 | $6.00 | |
| 50 | Kimi K2.5 Moonshot AI | 1249 | 58.5 | $2.50 | |
| 51 | GLM 5 Z.ai | 1248 | 59.7 | $1.92 | |
| 52 | Claude Opus 4.5 Anthropic | 1247 | 66.8 | $25.00 | |
| 53 | MiniMax M2.7 MiniMax | 1243 | 28.4 | $1.20 | |
| 54 | Qwen3.6 Plus Qwen | 1243 | 37.4 | $1.95 | |
| 55 | DeepSeek V4 Flash 0731 DeepSeek | 1238 | Not benchmarked | $1.32 | |
| 56 | DeepSeek V4 Pro 0423 DeepSeek | 1235 | 35.3 | $1.9105 | |
| 57 | Gemini 3.1 Pro Preview Google | 1234 | 35.3 | $12.00 | |
| 58 | Grok 4.20 xAI | 1231 | Not benchmarked | $2.50 | |
| 59 | Hy3 Tencent | 1231 | 43.9 | $0.528 | |
| 60 | GLM 5V Turbo Z.ai | 1230 | 28.8 | $4.00 | |
| 61 | GLM 4.7 Z.ai | 1226 | 33.6 | $1.98 | |
| 62 | T Inkling Thinking Machines | 1224 | 49.7 | $4.05 | |
| 63 | MiniMax M2.5 MiniMax | 1222 | 59.4 | $0.95 | |
| 64 | DeepSeek V4 Flash 0423 DeepSeek | 1208 | 54.6 | $0.182 | |
| 65 | MiniMax M2.1 MiniMax | 1202 | Not benchmarked | $1.20 | |
| 66 | GPT-5.4 OpenAI | 1199 | 37.7 | $15.00 | |
| 67 | Gemini 3 Flash Preview Google | 1196 | 75.8 | $3.00 | |
| 68 | GPT-5.2 OpenAI | 1196 | 58.9 | $14.00 | |
| 69 | Grok 4.3 xAI | 1196 | 22.5 | $2.50 | |
| 70 | Step 3.7 Flash StepFun | 1196 | Not benchmarked | $1.15 | |
| 71 | GLM 4.7 Flash Z.ai | 1194 | Not benchmarked | $0.40 | |
| 72 | Qwen3.5 397B A17B Qwen | 1191 | 28.6 | $2.34 | |
| 73 | Claude Sonnet 4.5 Anthropic | 1190 | 59.1 | $15.00 | |
| 74 | DeepSeek V3.1 Terminus DeepSeek | 1188 | Not benchmarked | $1.00 | |
| 75 | 1188 | Not benchmarked | $1.56 | ||
| 76 | GPT-5.1 OpenAI | 1187 | 53.3 | $10.00 | |
| 77 | GPT-5 OpenAI | 1185 | 72.2 | $10.00 | |
| 78 | DeepSeek V3.2 Exp DeepSeek | 1179 | 55.0 | $0.41 | |
| 79 | Claude Opus 4.1 Anthropic | 1177 | 27.2 | $75.00 | |
| 80 | Solar Pro 4 Upstage | 1177 | 24.4 | $0.60 | |
| 81 | DeepSeek V3.2 DeepSeek | 1175 | 54.4 | $0.42 | |
| 82 | GLM 4.6 Z.ai | 1174 | 52.2 | $2.20 | |
| 83 | GLM 4.5 Z.ai | 1170 | 64.2 | $2.20 | |
| 84 | Nemotron 3.5 Super VL NVIDIA | 1169 | 65.5 | $0.50 | |
| 85 | Gemini 2.5 Pro Google | 1167 | 37.0 | $10.00 | |
| 86 | GPT-5.3-Codex OpenAI | 1163 | 29.5 | $14.00 | |
| 87 | Mistral Large 3 2512 Mistral AI | 1162 | Not benchmarked | $1.50 | |
| 88 | GPT-5.1-Codex OpenAI | 1162 | 50.5 | $10.00 | |
| 89 | Solar Mini 4 Upstage | 1161 | Not benchmarked | $0.20 | |
| 90 | 1159 | 50.0 | $1.00 | ||
| 91 | R1 0528 DeepSeek | 1149 | 71.4 | $2.15 | |
| 92 | GLM 4.5 Air Z.ai | 1147 | Not benchmarked | $0.85 | |
| 93 | Nemotron 3 Ultra NVIDIA | 1145 | Not benchmarked | $2.40 | |
| 94 | MiniMax M2 MiniMax | 1144 | 45.3 | $1.02 | |
| 95 | Mistral Medium 3.5 Mistral AI | 1142 | 9.2 | $7.50 | |
| 96 | Trinity Large Thinking Arcee AI | 1137 | 5.4 | $0.80 | |
| 97 | Mistral Medium 3.1 Mistral AI | 1133 | Not benchmarked | $2.00 | |
| 98 | GPT-5 Mini OpenAI | 1125 | 59.8 | $2.00 | |
| 99 | DeepSeek V3.1 DeepSeek | 1124 | Not benchmarked | $0.95 | |
| 100 | Claude Haiku 4.5 Anthropic | 1122 | 49.8 | $5.00 |
Showing the first 100 of 131. Every row, with its source date, is in the full dataset.
What this measures
When people compare two websites built from the same prompt, whose do they prefer?
Design Arena shows people two designs made by two models from the same prompt (a website) and asks which is better. The number is the Elo rating built from those votes, as Design Arena publishes it.
Read from Design Arena on October 20, 2026. Each model appears once with its best published setting; rows that match no model in our catalog are left out. Design Arena: websites benchmark page