Best AI model for websites
In websites, work within 60 Elo of the best model is available for 4× less: MiMo-V2.6-Pro at $0.472 per thousand tasks against $1.98 for Muse Spark 1.3.
118 models rated head to head in this category, not scored by a general benchmark and assumed to transfer. Models are given a brief and build a complete page covering layout, copy, styling and responsiveness, then judged against each other by human preference.
Nothing cheaper is as good
7 models are undominated here: for each, no cheaper model scores within 18 Elo. Everything else in the table below is beaten on both counts at once.
- 1gpt-oss-20b857 Elo$0.066 / 1k131K contextfp4 on some hosts
- 2Ministral 3 3B 25121032 Elo$0.085 / 1k131K context
- 3Solar Mini 41179 Elo$0.088 / 1k524K context
- 4DeepSeek V4 Flash 04231211 Elo$0.170 / 1k1.05M contextfp4 on some hosts
- 5MiMo-V2.51270 Elo$0.183 / 1k1.05M context
- 6MiMo-V2.6-Pro1312 Elo$0.472 / 1k1.05M context
- 7Muse Spark 1.31361 Elo$1.98 / 1k1.05M context
Every model rated on websites
Elo measured in this category specifically. A model strong overall can sit well down this table, which is the whole reason for ranking by task.
| # | Model | Elo | $/1k | Context | Hosts |
|---|---|---|---|---|---|
| 1 | Muse Spark 1.3 | 1361 | $1.98 | 1.05M | 1 |
| 2 | Kimi K3 | 1339 | $6.24 | 1.05M | 25 |
| 3 | Claude Opus 5.5 | 1335 | $8.86 | 1.00M | 11 |
| 4 | Claude Sonnet 5.5 | 1327 | $4.03 | 1.00M | 8 |
| 5 | GPT-6.1 Sol | 1323 | $4.43 | 1.05M | 8 |
| 6 | Claude Fable 5.1 | 1317 | $19.98 | 1.00M | 4 |
| 7 | Muse Spark 1.2 | 1317 | $1.98 | 1.05M | 1 |
| 8 | Claude Opus 5 | 1313 | $11.25 | 1.00M | 11 |
| 9 | MiMo-V2.6-Pro | 1312 | $0.472 | 1.05M | 4 |
| 10 | Gemini 3.7 Flash | 1308 | $1.54 | 1.05M | 6 |
| 11 | Gemini 3.8 Flash | 1305 | $1.54 | 1.05M | 6 |
| 12 | Gemini 3.6 Flash | 1303 | $1.54 | 1.05M | 7 |
| 13 | GLM 5.3 | 1302 | $1.93 | 1.05M | 41 |
| 14 | Claude Fable 5 | 1300 | $20.52 | 1.00M | 6 |
| 15 | Claude Opus 4.6 | 1295 | $10.26 | 1.00M | 6 |
| 16 | Grok 4.6 | 1293 | $3.40 | 500K | 6 |
| 17 | GLM 5.2 | 1290 | $2.00 | 1.05M | 34 |
| 18 | GLM 5.1 | 1290 | $2.11 | 203K | 13 |
| 19 | Claude Sonnet 4.6 | 1289 | $6.16 | 1.00M | 9 |
| 20 | Grok 4.5 | 1286 | $4.36 | 500K | 4 |
| 21 | Claude Sonnet 5 | 1279 | $4.30 | 1.00M | 10 |
| 22 | Qwen3.7 Max | 1277 | $2.61 | 1.00M | 2 |
| 23 | GLM 5.3 Flash | 1276 | $0.230 | 1.05M | 32 |
| 24 | Muse Spark 1.1 | 1273 | $1.98 | 1.05M | 1 |
| 25 | Qwen3.7 Plus | 1273 | $0.584 | 1.00M | 2 |
| 26 | Kimi K2.6 | 1273 | $1.52 | 262K | 17 |
| 27 | Claude Haiku 5.5 | 1272 | $0.215 | 1.00M | 10 |
| 28 | MiMo-V2.5-Pro | 1271 | $0.497 | 1.05M | 6 |
| 29 | Gemini 3.5 Flash | 1270 | $3.53 | 1.05M | 7 |
| 30 | MiMo-V2.5 | 1270 | $0.183 | 1.05M | 5 |
| 31 | Kimi K2.7 Code | 1268 | $1.75 | 262K | 12 |
| 32 | GLM 5 Turbo | 1267 | $1.95 | 203K | 1 |
| 33 | MiniMax M3 | 1262 | $0.547 | 1.05M | 13 |
| 34 | Claude Opus 4.8 | 1261 | $11.25 | 1.00M | 11 |
| 35 | GPT-5.5 | 1261 | $12.90 | 1.05M | 7 |
| 36 | Mistral Large 4 | 1260 | $1.00 | 1.05M | 3 |
| 37 | Grok 4.7 | 1258 | $3.40 | 500K | 5 |
| 38 | Gemini 3.1 Pro Preview | 1257 | $4.70 | 1.05M | 6 |
| 39 | GLM 5 | 1252 | $1.42 | 203K | 8 |
| 40 | Kimi K2.5 | 1252 | $1.22 | 262K | 5 |
| 41 | Claude Opus 4.5 | 1251 | $10.26 | 200K | 6 |
| 42 | Qwen3.6 Plus | 1246 | $0.975 | 1.00M | 1 |
| 43 | MiniMax M2.7 | 1246 | $0.534 | 197K | 6 |
| 44 | DeepSeek V4 Flash 0731 | 1242 | $0.233 | 1.05M | 27 |
| 45 | DeepSeek V4 Pro 0423 | 1239 | $1.68 | 1.05M | 15 |
| 46 | GLM 5V Turbo | 1234 | $1.95 | 203K | 1 |
| 47 | Grok 4.20 | 1234 | $2.17 | 2.00M | 4 |
| 48 | GLM 4.7 | 1229 | $0.980 | 203K | 6 |
| 49 | Inkling | 1228 | $1.81 | 524K | 2 |
| 50 | MiniMax M2.5 | 1227 | $0.523 | 197K | 9 |
| 51 | GPT-5.4 | 1224 | $6.45 | 1.05M | 7 |
| 52 | DeepSeek V4 Flash 0423 | 1211 | $0.170 | 1.05M | 16 |
| 53 | MiniMax M2.1 | 1206 | $0.706 | 205K | 2 |
| 54 | Gemini 3 Flash Preview | 1201 | $1.18 | 1.05M | 6 |
| 55 | Grok 4.3 | 1200 | $2.17 | 1.00M | 4 |
| 56 | Step 3.7 Flash | 1199 | $0.470 | 262K | 2 |
| 57 | GLM 4.7 Flash | 1199 | $0.193 | 128K | 3 |
| 58 | GPT-5.2 | 1199 | $5.17 | 400K | 4 |
| 59 | Qwen3.5 397B A17B | 1195 | $1.49 | 262K | 10 |
| 60 | Claude Sonnet 4.5 | 1193 | $6.16 | 1.00M | 7 |
| 61 | Qwen3.5 Plus 2026-02-15 | 1191 | $0.780 | 1.00M | 1 |
| 62 | GPT-5.1 | 1191 | $3.69 | 400K | 5 |
| 63 | DeepSeek V3.1 Terminus | 1191 | $0.672 | 164K | 2 |
| 64 | GPT-5 | 1189 | $3.69 | 400K | 3 |
| 65 | Hy3 | 1185 | $0.265 | 262K | 5 |
| 66 | DeepSeek V3.2 Exp | 1183 | $0.447 | 164K | 1 |
| 67 | Solar Pro 4 | 1181 | $0.164 | 524K | 2 |
| 68 | Claude Opus 4.1 | 1181 | $30.78 | 200K | 1 |
| 69 | Solar Mini 4 | 1179 | $0.088 | 524K | 2 |
| 70 | DeepSeek V3.2 | 1179 | $0.504 | 164K | 12 |
| 71 | GLM 4.6 | 1179 | $0.954 | 203K | 4 |
| 72 | GLM 4.5 | 1174 | $1.03 | 131K | 1 |
| 73 | Gemini 2.5 Pro | 1170 | $3.69 | 1.05M | 7 |
| 74 | GPT-5.3-Codex | 1167 | $5.17 | 400K | 3 |
| 75 | Mistral Large 3 2512 | 1167 | $0.760 | 262K | 2 |
| 76 | GPT-5.1-Codex | 1166 | $3.69 | 400K | 1 |
| 77 | Qwen3 Coder 480B A35B | 1163 | $0.666 | 262K | 3 |
| 78 | R1 0528 | 1154 | $1.18 | 164K | 4 |
| 79 | GLM 4.5 Air | 1151 | $0.343 | 131K | 3 |
| 80 | Claude Sonnet 4 | 1150 | $6.16 | 200K | 2 |
| 81 | MiniMax M2 | 1147 | $0.666 | 205K | 2 |
| 82 | Nemotron 3 Ultra | 1145 | $1.09 | 203K | 4 |
| 83 | Trinity Large Thinking | 1141 | $0.403 | 262K | 1 |
| 84 | Mistral Medium 3.1 | 1137 | $0.821 | 131K | 3 |
| 85 | GPT-5 Mini | 1129 | $0.738 | 400K | 4 |
| 86 | Claude Haiku 4.5 | 1127 | $2.15 | 200K | 8 |
| 87 | DeepSeek V3.1 | 1127 | $0.808 | 164K | 5 |
| 88 | DeepSeek V3 | 1124 | $0.634 | 128K | 2 |
| 89 | Kimi K2 Thinking | 1118 | $1.15 | 262K | 2 |
| 90 | Gemini 2.5 Flash | 1118 | $0.916 | 1.05M | 7 |
| 91 | GPT-5.1-Codex-Mini | 1115 | $0.742 | 400K | 1 |
| 92 | Kimi K2 0905 | 1112 | $1.47 | 262K | 1 |
| 93 | GPT-5 Nano | 1106 | $0.148 | 400K | 4 |
| 94 | Qwen3 Coder 30B A3B Instruct | 1092 | $0.264 | 262K | 2 |
| 95 | Gemini 3.1 Flash Lite Preview | 1087 | $0.588 | 1.05M | 3 |
| 96 | Ministral 3 14B 2512 | 1086 | $0.170 | 262K | 3 |
| 97 | Mistral Medium 3 | 1083 | $0.821 | 131K | 3 |
| 98 | Ministral 3 8B 2512 | 1068 | $0.128 | 262K | 3 |
| 99 | Qwen3 235B A22B Instruct 2507 | 1062 | $0.405 | 262K | 7 |
| 100 | Qwen3 235B A22B Thinking 2507 | 1057 | $1.59 | 128K | 1 |
| 101 | Kimi K2 0711 | 1055 | $1.37 | 131K | 1 |
| 102 | GPT-4.1 | 1043 | $3.72 | 1.05M | 3 |
| 103 | o3 | 1041 | $3.72 | 200K | 1 |
| 104 | Mercury 2 | 1034 | $0.363 | 128K | 1 |
| 105 | Ministral 3 3B 2512 | 1032 | $0.085 | 131K | 3 |
| 106 | Codestral 2508 | 1017 | $0.456 | 256K | 2 |
| 107 | GPT-4.1 Mini | 1002 | $0.744 | 1.05M | 3 |
| 108 | o4 Mini | 990 | $2.05 | 200K | 1 |
| 109 | GPT-4.1 Nano | 977 | $0.186 | 1.05M | 3 |
| 110 | gpt-oss-120b | 972 | $0.285 | 131K | 24 |
| 111 | Qwen3 30B A3B | 959 | $0.294 | 41K | 1 |
| 112 | Mistral Small 3.2 24B | 900 | $0.193 | 128K | 4 |
| 113 | Llama 4 Maverick | 875 | $0.462 | 524K | 4 |
| 114 | gpt-oss-20b | 857 | $0.066 | 131K | 11 |
| 115 | Nova Premier 1.0 | 840 | $5.40 | 1.00M | 1 |
| 116 | GPT-4o | 836 | $5.10 | 128K | 2 |
| 117 | Nova Pro 1.0 | 801 | $1.92 | 300K | 2 |
| 118 | Llama 4 Scout | 755 | $0.393 | 328K | 3 |
Other tasks
Elo comes from head-to-head build-offs via OpenRouter, compiled 12 October 2026 and refreshed daily. Cost is the median across every host serving each model, priced on 1,200 input and 300 output tokens with 60% cached. See the findings for method, or download the data.