Website development rankings
Best AI for web development
Compare current website-building results from head-to-head Design Arena tournaments. ELO is a rating that rises when a model wins against strong competitors.
Current source snapshot
Jul 17, 2026
140 ranked entries
Live category evidence
Current Design Arena leaderboard
| Rank | Model | ELO | Win rate | Average time | Blended price | Tournament runs |
|---|---|---|---|---|---|---|
| 1 |
GLM 5.2
Zhipu
|
1,343 | 60.9% | 301.3s | $2.15 | 11,176 |
| 2 |
Claude Fable 5
Anthropic
|
1,335 | 62.0% | 157.1s | $22 | 8,663 |
| 3 |
Claude Opus 4.6
Anthropic
|
1,328 | 62.9% | 204.3s | $11 | 19,598 |
| 4 |
Claude Opus 4.7
Anthropic
|
1,327 | 60.2% | 161.7s | $11 | 19,389 |
| 5 |
Grok 4.5
xAI
|
1,327 | 58.2% | 137.7s | $3 | 5,992 |
| 6 |
Claude Opus 4.6 (Thinking)
Anthropic
|
1,320 | 61.7% | 207.3s | $11 | 11,862 |
| 7 |
Claude Sonnet 4.6
Anthropic
|
1,318 | 61.3% | 191.3s | $6 | 20,153 |
| 8 |
Claude Sonnet 5
Anthropic
|
1,314 | 57.6% | 145.3s | $4 | 4,936 |
| 10 |
Kimi K2.6
Moonshot
|
1,309 | 56.3% | 462.4s | $1.71 | 28,120 |
| 11 |
Kimi K2.7 Code
Moonshot
|
1,305 | 56.0% | 341.1s | $1.71 | 10,711 |
| 12 |
GLM 5 Turbo
Zhipu
|
1,303 | 56.2% | 338.1s | $1.9 | 12,847 |
| 13 |
Muse Spark 1.1
Meta
|
1,301 | 55.1% | 187.8s | $2 | 10,843 |
| 14 |
MiMo-V2.5-Pro
Xiaomi
|
1,301 | 56.0% | 346.5s | $1.5 | 5,024 |
| 15 |
Qwen3.7 Max
Alibaba
|
1,296 | 56.2% | 282.4s | $2.21 | 12,830 |
| 16 |
MiMo-V2.5
Xiaomi
|
1,293 | 55.0% | 204.7s | $0.14 | 14,447 |
| 17 |
MiniMax M3
MiniMax
|
1,291 | 55.0% | 218.2s | $0.53 | 7,186 |
| 18 |
GLM 5.1
Zhipu
|
1,290 | 55.1% | 302.8s | $2.14 | 5,939 |
| 19 |
Qwen3.7 Plus
Alibaba
|
1,290 | 53.1% | 365.9s | $0.56 | 848 |
| 20 |
Gemini 3.5 Flash
Google
|
1,287 | 56.2% | 76.2s | $3.38 | 28,872 |
| 21 |
GPT-5.5
OpenAI
|
1,285 | 55.8% | 154.1s | $11.25 | 22,204 |
| 22 |
Gemini 3 Pro Preview
Google
|
1,284 | 64.3% | 90.7s | $4.5 | 12,275 |
| 23 |
Kimi K2.5 (Thinking)
Moonshot
|
1,282 | 55.8% | 245.6s | $1.3 | 27,045 |
| 24 |
Claude Opus 4.5
Anthropic
|
1,280 | 59.7% | 136.2s | $10 | 18,995 |
| 25 |
GLM 5
Zhipu
|
1,280 | 55.0% | 233.0s | $1.55 | 31,968 |
| 26 |
MiniMax M2.7
MiniMax
|
1,278 | 54.2% | 216.9s | $0.53 | 14,613 |
| 27 |
Claude Opus 4.8
Anthropic
|
1,273 | 54.3% | 89.6s | $10 | 15,466 |
| 28 |
DeepSeek-V4-Pro
DeepSeek
|
1,267 | 53.0% | 301.2s | $2.18 | 10,218 |
| 29 |
Gemini 3.1 Pro Preview
Google
|
1,265 | 56.5% | 103.9s | $4.5 | 36,203 |
| 30 |
GLM 5V Turbo
Zhipu
|
1,261 | 50.9% | 214.9s | $1.9 | 29,673 |
| 31 |
Grok 4.20 Beta (Reasoning)
xAI
|
1,260 | 54.5% | 96.9s | $1.56 | 16,956 |
| 32 |
GLM 4.7
Zhipu
|
1,258 | 55.2% | 172.8s | $1 | 29,267 |
| 33 |
MiniMax M2.5
MiniMax
|
1,255 | 57.5% | 147.5s | $0.44 | 5,389 |
| 34 |
GPT-5.4 (Design Skill, Medium)
OpenAI
|
1,252 | 52.6% | 163.4s | $5.63 | 4,832 |
| 35 |
Qwen3.6 Plus
Alibaba
|
1,252 | 51.0% | 311.5s | $0.73 | 16,468 |
| 36 |
GPT-5.4 (Medium)
OpenAI
|
1,248 | 52.8% | 158.0s | $5.63 | 10,752 |
| 37 |
Grok 4.20 Beta
xAI
|
1,246 | 52.8% | 77.6s | $1.56 | 17,286 |
| 38 |
DeepSeek-V4-Flash
DeepSeek
|
1,240 | 50.2% | 137.8s | $0.17 | 11,320 |
| 39 |
Nex N2 Pro
Nex Agi
|
1,240 | 46.6% | 426.4s | $0.44 | 3,681 |
| 40 |
Claude 3.7 Sonnet
Anthropic
|
1,236 | 58.3% | 121.8s | — | 7,529 |
| 41 |
MiniMax M2.1
MiniMax
|
1,234 | 55.4% | 122.7s | $0.53 | 9,381 |
| 42 |
GPT-5.4 (None)
OpenAI
|
1,231 | 50.8% | 79.7s | $5.63 | 13,537 |
| 43 |
GPT-5.4 (Low)
OpenAI
|
1,230 | 50.5% | 89.4s | $5.63 | 11,777 |
| 44 |
Gemini 3 Flash Preview
Google
|
1,228 | 57.0% | 35.8s | $1.13 | 9,357 |
| 45 |
Grok 4.3
xAI
|
1,228 | 48.7% | 82.7s | $1.56 | 29,275 |
| 46 |
GPT-5.2 (XHigh)
OpenAI
|
1,227 | 54.5% | 239.4s | $4.81 | 380 |
| 47 |
GLM 4.7 Flash
Zhipu
|
1,227 | 54.0% | 159.0s | $0.15 | 5,798 |
| 48 |
Qwen3.5 397B A17B
Alibaba
|
1,223 | 52.6% | 140.7s | $1.09 | 4,149 |
| 49 |
Claude Sonnet 4.5
Anthropic
|
1,222 | 52.9% | 106.2s | $6.6 | 21,320 |
| 50 |
Claude Sonnet 4.5 (Thinking)
Anthropic
|
1,222 | 52.8% | 100.1s | $6.6 | 20,899 |
| 51 |
GPT-5.1 (High)
OpenAI
|
1,220 | 54.1% | 170.0s | $3.44 | 8,405 |
| 52 |
Qwen3.5 Plus 02-15
Alibaba
|
1,220 | 50.1% | 140.2s | $0.59 | 9,623 |
| 53 |
DeepSeek-V3.1 (Thinking)
DeepSeek
|
1,219 | 56.4% | 199.8s | $0.44 | 7,784 |
| 54 |
GPT-5 (Minimal)
OpenAI
|
1,217 | 53.7% | 118.3s | $3.44 | 17,332 |
| 55 |
GPT-5.2 (None)
OpenAI
|
1,217 | 51.3% | 176.2s | $4.81 | 16,146 |
| 56 |
Hy3
Tencent
|
1,216 | 43.6% | 255.8s | $0.25 | 676 |
| 57 |
GPT-5.2 (Medium)
OpenAI
|
1,216 | 50.9% | 183.9s | $4.81 | 15,472 |
| 58 |
GPT-5 (High)
OpenAI
|
1,215 | 55.3% | 165.1s | $3.44 | 6,807 |
| 59 |
Step 3.7 Flash
Stepfun
|
1,215 | 46.3% | 101.0s | $0.44 | 8,165 |
| 60 |
DeepSeek-V3.2-Exp
DeepSeek
|
1,211 | 54.2% | 203.3s | $0.31 | 9,583 |
| 61 |
GPT-5.1 (Medium)
OpenAI
|
1,211 | 53.0% | 144.7s | $3.44 | 10,709 |
| 62 |
GPT-5.2 (Low)
OpenAI
|
1,210 | 51.0% | 175.5s | $4.81 | 13,533 |
| 63 |
Claude Opus 4.1 (Thinking)
Anthropic
|
1,209 | 55.3% | 107.2s | $30 | 7,473 |
| 64 |
MiMo-V2-Flash
Xiaomi
|
1,209 | 49.9% | 65.9s | — | 16,624 |
| 65 |
DeepSeek-V3.2
DeepSeek
|
1,207 | 50.1% | 201.8s | $1 | 14,620 |
| 66 |
GLM 4.6
Zhipu
|
1,207 | 54.4% | 115.1s | $0.96 | 14,578 |
| 67 |
GPT-5.1 (Low)
OpenAI
|
1,204 | 52.4% | 122.9s | $3.44 | 10,832 |
| 68 |
GLM 4.5
Zhipu
|
1,202 | 53.8% | 147.8s | $1 | 9,161 |
| 69 |
Gemini 2.5 Pro
Google
|
1,200 | 58.8% | 75.8s | $3.44 | 3,845 |
| 70 |
GPT-5.1 (None)
OpenAI
|
1,200 | 51.7% | 113.9s | $3.44 | 10,899 |
| 71 |
Claude Opus 4
Anthropic
|
1,197 | 54.6% | 91.0s | $30 | 7,932 |
| 72 |
Claude Opus 4.1
Anthropic
|
1,197 | 49.4% | 112.0s | $30 | 20,469 |
| 73 |
GPT-5.3 Codex
OpenAI
|
1,196 | 48.7% | 52.3s | $4.81 | 11,788 |
| 74 |
GPT-5.1 Codex
OpenAI
|
1,194 | 56.0% | 73.1s | $3.44 | 1,931 |
| 75 |
Mistral Large 3 (2512)
Mistral
|
1,194 | 49.4% | 122.5s | $0.75 | 14,978 |
| 76 |
Hy3 Low
Tencent
|
1,193 | 39.8% | 98.7s | $0.25 | 1,187 |
| 77 |
Qwen3 Coder 480B A35B Instruct
Alibaba
|
1,191 | 61.7% | 68.3s | $0.67 | 902 |
| 78 |
GPT-5.2 (High)
OpenAI
|
1,186 | 49.5% | 152.4s | $4.81 | 1,661 |
| 79 |
DeepSeek-R1-0528
DeepSeek
|
1,182 | 52.7% | 89.8s | $1.15 | 8,357 |
| 80 |
GLM 4.5 Air
Zhipu
|
1,179 | 51.3% | 126.5s | $0.43 | 8,103 |
| 81 |
Claude Sonnet 4
Anthropic
|
1,178 | 52.4% | 74.8s | $6 | 8,318 |
| 82 |
MiniMax M2 Stable
MiniMax
|
1,175 | 48.0% | 124.3s | $0.45 | 5,275 |
| 83 |
Claude Sonnet 4 (Thinking)
Anthropic
|
1,172 | 50.6% | 73.5s | $6 | 7,680 |
| 84 |
Trinity Large Thinking
Arcee Ai
|
1,167 | 41.3% | 96.6s | $0.39 | 6,482 |
| 85 |
Mistral Medium 3.1 (2508)
Mistral
|
1,165 | 46.0% | 124.9s | $0.8 | 14,173 |
| 86 |
Prime Intellect. INTELLECT-3
Prime Intellect
|
1,159 | 42.9% | 83.7s | — | 15,077 |
| 87 |
GPT-5 mini (Default)
OpenAI
|
1,157 | 44.3% | 133.8s | $0.69 | 19,371 |
| 88 |
Grok 4 Fast
xAI
|
1,157 | 44.0% | 50.1s | — | 18,019 |
| 89 |
DeepSeek-V3.1
DeepSeek
|
1,155 | 48.0% | 47.9s | $0.46 | 9,535 |
| 90 |
Claude Haiku 4.5
Anthropic
|
1,154 | 45.0% | 52.7s | $2.2 | 18,482 |
| 91 |
DeepSeek-V3-0324
DeepSeek
|
1,153 | 48.5% | 79.6s | $0.63 | 9,090 |
| 92 |
Qwen3 Max
Alibaba
|
1,151 | 44.4% | 114.1s | $1.56 | 16,354 |
| 93 |
Gemini 2.5 Flash Preview 09-2025
Google
|
1,148 | 47.1% | 48.6s | $0.85 | 9,097 |
| 94 |
Grok 4.1 Fast
xAI
|
1,147 | 43.0% | 37.5s | — | 16,318 |
| 95 |
Grok 4.1 Fast (Reasoning)
xAI
|
1,147 | 42.8% | 60.8s | — | 16,053 |
| 96 |
Kimi K2 Thinking
Moonshot
|
1,146 | 48.8% | 245.1s | $1.08 | 503 |
| 97 |
GPT-5.1 Codex Mini
OpenAI
|
1,143 | 42.8% | 46.5s | $0.69 | 17,161 |
| 98 |
Kimi K2 0905 Preview
Moonshot
|
1,140 | 48.3% | 193.4s | $1.08 | 672 |
| 99 |
GPT-5 nano (Default)
OpenAI
|
1,134 | 48.9% | 90.8s | $0.14 | 3,043 |
| 100 |
Gemini 2.5 Flash Lite Preview 09-2025
Google
|
1,132 | 48.1% | 13.7s | — | 3,139 |
| 101 |
Nemotron 3 Ultra
NVIDIA
|
1,131 | 32.6% | 154.4s | $0 | 4,868 |
| 102 |
Kimi K2 Instruct 0905
Moonshot
|
1,121 | 54.5% | 33.0s | $1.08 | 162 |
| 103 |
Qwen3 Coder 30B A3B Instruct
Alibaba
|
1,120 | 57.1% | 29.9s | $0.12 | 492 |
| 104 |
Gemini 3.1 Flash-Lite Preview
Google
|
1,114 | 36.5% | 6.4s | $0.56 | 12,910 |
| 105 |
Ministral 3 14B (2512)
Mistral
|
1,114 | 44.8% | 84.8s | $0.35 | 1,017 |
| 106 |
Mistral Medium 3 (2505)
Mistral
|
1,112 | 47.7% | 78.3s | $0.8 | 2,890 |
| 107 |
Gemini 2.5 Flash
Google
|
1,107 | 45.7% | 77.9s | $0.85 | 3,254 |
| 108 |
Grok 3
xAI
|
1,101 | 39.7% | 47.9s | — | 13,192 |
| 109 |
Grok 4 Fast (Reasoning)
xAI
|
1,099 | 36.6% | 27.4s | — | 18,584 |
| 110 |
Ministral 3 8B (2512)
Mistral
|
1,098 | 42.9% | 56.1s | $0.3 | 1,034 |
| 111 |
Qwen3-235B-A22B-Instruct-2507
Alibaba
|
1,091 | 43.7% | 55.1s | $0.3 | 3,206 |
| 112 |
Qwen3-235B-A22B-Thinking-2507
Alibaba
|
1,085 | 42.1% | 113.1s | $0.75 | 2,908 |
| 113 |
Kimi K2
Moonshot
|
1,083 | 53.1% | 59.8s | $1 | 596 |
| 114 |
GPT-4.1
OpenAI
|
1,071 | 52.3% | 71.5s | $3.5 | 767 |
| 115 |
o3
OpenAI
|
1,069 | 53.8% | 53.0s | $3.5 | 590 |
| 116 |
Qwen3-235B-A22B
Alibaba
|
1,063 | 40.5% | 76.4s | $0.8 | 2,399 |
| 117 |
Devstral Medium
Mistral
|
1,062 | 40.4% | 25.6s | — | 3,285 |
| 118 |
Ministral 3 3B (2512)
Mistral
|
1,060 | 38.2% | 37.6s | $0.15 | 1,253 |
| 119 |
Grok 4
xAI
|
1,056 | 33.7% | 63.4s | — | 11,878 |
| 120 |
Codestral 2508
Mistral
|
1,045 | 37.8% | 18.7s | $0.45 | 3,142 |
| 121 |
Grok Code Fast 1
xAI
|
1,038 | 35.3% | 16.7s | — | 8,460 |
| 122 |
GPT-4.1 mini
OpenAI
|
1,030 | 47.8% | 70.4s | $0.7 | 681 |
| 123 |
Olmo 3.1 32B Think
Allenai
|
1,021 | 26.4% | 130.0s | — | 8,236 |
| 124 |
Mercury 2
Inception
|
1,020 | 20.2% | 5.1s | $0.38 | 3,857 |
| 125 |
o4-mini
OpenAI
|
1,018 | 47.1% | 40.6s | $1.93 | 877 |
| 126 |
GPT-4.1 nano
OpenAI
|
1,005 | 48.1% | 27.3s | $0.18 | 858 |
| 127 |
GPT OSS 120B
OpenAI
|
1,000 | 32.5% | 25.4s | $0.26 | 2,487 |
| 128 |
Qwen3 30B-A3B
Alibaba
|
987 | 37.7% | 29.1s | $0.24 | 1,184 |
| 129 |
Llama 3.1 Nemotron Ultra 253B
NVIDIA
|
978 | 26.7% | 33.0s | — | 1,464 |
| 130 |
Grok 3 Mini
xAI
|
972 | 30.6% | 24.4s | — | 3,594 |
| 131 |
Qwen3 30B-A3B Thinking 2507
Alibaba
|
963 | 35.5% | 41.3s | $0.49 | 377 |
| 132 |
Mistral Small 3.2
Mistral
|
928 | 38.3% | 18.3s | $0.14 | 567 |
| 133 |
Llama 4 Maverick
Meta Llama
|
904 | 34.4% | 13.6s | $0.35 | 790 |
| 134 |
Mistral Large 2.1 (2411)
Mistral
|
900 | 33.0% | 33.3s | — | 584 |
| 135 |
GPT OSS 20B
OpenAI
|
885 | 27.9% | 13.1s | $0.08 | 471 |
| 136 |
Amazon Nova Premier
Amazon
|
868 | 26.2% | 14.7s | $5 | 269 |
| 137 |
GPT-4o
OpenAI
|
864 | 31.5% | 14.2s | $4.38 | 854 |
| 138 |
Devstral Small 1.1
Mistral
|
845 | 27.1% | 6.8s | — | 600 |
| 139 |
Codestral 2 (2501)
Mistral
|
842 | 28.3% | 6.1s | $0.45 | 702 |
| 140 |
Amazon Nova Pro
Amazon
|
829 | 21.4% | 14.8s | $1.4 | 365 |
| 141 |
Llama 4 Scout
Meta Llama
|
783 | 22.7% | 11.9s | $0.15 | 615 |
What this ranking means. ELO summarizes head-to-head preference results inside this category. It is useful evidence for the named task, but it remains separate from the broader Vibe Coding Index.