Skip to content

Benchmark

Design Arena Elo (models/website)

People compare model answers and choose which they prefer. Higher scores are better.

Models measured
148
Scores recorded
161
Sources
1
IQ weight
1.85

Score distribution

1,153.6
Mean
1,172
Median
128.5
Std dev
983
10th percentile
1,300
90th percentile
737
Min
7371,374
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.85 for chat models, measured over 148 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
101Solar Mini 4· Design Arena Leaderboards data1,134
102Claude Sonnet 4· Design Arena Leaderboards data1,133
103v0-1.5-lg· Design Arena Leaderboards data1,133
104Mistral Medium 3.5· Design Arena Leaderboards data1,129
105Claude Sonnet 4thinking· Design Arena Leaderboards data1,126
106Nemotron 3 Ultra· Design Arena Leaderboards data1,125
107Trinity Large Thinking· Design Arena Leaderboards data1,124
108Mistral Medium 3.1· Design Arena Leaderboards data1,120
109GPT-5 mini· Design Arena Leaderboards data1,112
110DeepSeek V3.1· Design Arena Leaderboards data1,110
111Claude Haiku 4.5· Design Arena Leaderboards data1,109
112DeepSeek Chat· Design Arena Leaderboards data1,107
113Qwen3 Max· Design Arena Leaderboards data1,106
114INTELLECT-3· Design Arena Leaderboards data1,104
115gemini-2.5-flash-preview-09-2025· Design Arena Leaderboards data1,102
116Grok 4 Fastnon-reasoning· Design Arena Leaderboards data1,101
117Kimi K2 Thinking· Design Arena Leaderboards data1,100
118GPT-5.1 Codex Mini· Design Arena Leaderboards data1,098
119Kimi K2 0905· Design Arena Leaderboards data1,094
120Grok 4.1 Fastnon-reasoning· Design Arena Leaderboards data1,092
121Grok 4.1 Fastreasoning· Design Arena Leaderboards data1,090
122GPT-5 nano· Design Arena Leaderboards data1,088
123Gemini 2.5 Flash Lite Preview 09-2025· Design Arena Leaderboards data1,086
124Kimi K2 Turbo Preview· Design Arena Leaderboards data1,077
125Qwen3 Coder 30B A3B Instruct· Design Arena Leaderboards data1,074
126Gemini 3.1 Flash Lite· Design Arena Leaderboards data1,070
127Ministral 14B2512· Design Arena Leaderboards data1,068
128Mistral Medium 3· Design Arena Leaderboards data1,066
129Gemini 2.5 Flash· Design Arena Leaderboards data1,061
130Ministral 8B2512· Design Arena Leaderboards data1,052
131v0-1.5-md· Design Arena Leaderboards data1,048
132Grok 3· Design Arena Leaderboards data1,046
133Qwen3 235B A22B Instruct 2507· Design Arena Leaderboards data1,045
134Grok 4 Fastreasoning· Design Arena Leaderboards data1,042
135Qwen 3 235B A22B Thinking 2507· Design Arena Leaderboards data1,040
136kimi-k2-0711-preview· Design Arena Leaderboards data1,037
137Magistral Medium· Design Arena Leaderboards data1,027
138GPT 4.1· Design Arena Leaderboards data1,026
139Qwen 3 235B A22B· Design Arena Leaderboards data1,018
140Ministral 3B2512· Design Arena Leaderboards data1,015
141Devstral Medium· Design Arena Leaderboards data1,007
142Mercury 2· Design Arena Leaderboards data1,005
143Grok 4· Design Arena Leaderboards data1,001
144Codestral 25.012508· Design Arena Leaderboards data1,000
145GPT-4.1 mini· Design Arena Leaderboards data984
146Grok Code Fast 1· Design Arena Leaderboards data983
147Magistral Small 2509· Design Arena Leaderboards data980
148Olmo 3.1 32B Think· Design Arena Leaderboards data965
149GPT-4.1 nano· Design Arena Leaderboards data960
150GPT OSS 120B· Design Arena Leaderboards data955

Showing 101–150 of 161 results.