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
51Grok 4.7· Design Arena Leaderboards data1,240
52Kimi K2.5· Design Arena Leaderboards data1,236
53GLM 5· Design Arena Leaderboards data1,235
54Claude Opus 4.5· Design Arena Leaderboards data1,234
55Qwen3.6 Plus· Design Arena Leaderboards data1,229
56Mercury 3· Design Arena Leaderboards data1,226
57DeepSeek V4 Flash 0731· Design Arena Leaderboards data1,225
58Hy4 Flash· Design Arena Leaderboards data1,225
59DeepSeek V4 Pro· Design Arena Leaderboards data1,222
60Gemini 3.1 Pro Preview· Design Arena Leaderboards data1,221
61GLM 5V Turbo· Design Arena Leaderboards data1,218
62Grok 4.20· Design Arena Leaderboards data1,216
63GLM 4.7· Design Arena Leaderboards data1,213
64Inkling FP4· Design Arena Leaderboards data1,210
65DeepSeek V4 Flash· Design Arena Leaderboards data1,195
66MiniMax M2.1· Design Arena Leaderboards data1,188
67GPT 5.4none· Design Arena Leaderboards data1,186
68GPT 5.4low· Design Arena Leaderboards data1,185
69Gemini 3 Flash· Design Arena Leaderboards data1,183
70Claude Sonnet 3.7· Design Arena Leaderboards data1,181
71GLM 4.7 Flash· Design Arena Leaderboards data1,181
72Grok 4.3· Design Arena Leaderboards data1,181
73Step 3.7 Flash· Design Arena Leaderboards data1,181
74Qwen3.5 397B A17B· Design Arena Leaderboards data1,178
75Claude Sonnet 4.5· Design Arena Leaderboards data1,177
76Claude Sonnet 4.5thinking· Design Arena Leaderboards data1,177
77Muse Glimmer 30B· Design Arena Leaderboards data1,175
78Qwen3.5 Plus· Design Arena Leaderboards data1,175
79DeepSeek V3.1thinking· Design Arena Leaderboards data1,174
80GPT 5.1high· Design Arena Leaderboards data1,174
81GPT 5· Design Arena Leaderboards data1,172
82GPT 5high· Design Arena Leaderboards data1,169
83Hy3 (Deep Infra)· Design Arena Leaderboards data1,169
84DeepSeek V3.2 Exp· Design Arena Leaderboards data1,166
85Claude Opus 4.1thinking, 20250805· Design Arena Leaderboards data1,164
86DeepSeek V3.2· Design Arena Leaderboards data1,162
87GLM 4.6· Design Arena Leaderboards data1,161
88GPT 5.1low· Design Arena Leaderboards data1,159
89GLM 4.5· Design Arena Leaderboards data1,157
90Gemini 2.5 Pro· Design Arena Leaderboards data1,154
91GPT 5.1none· Design Arena Leaderboards data1,154
92MiMo V2 Flash· Design Arena Leaderboards data1,154
93Claude Opus 4· Design Arena Leaderboards data1,152
94Claude Opus 4.1· Design Arena Leaderboards data1,152
95Nemotron 3.5 Super VL· Design Arena Leaderboards data1,151
96GPT-5.3 Codex· Design Arena Leaderboards data1,150
97GPT-5.1 Codex· Design Arena Leaderboards data1,149
98Mistral Large 3· Design Arena Leaderboards data1,149
99Qwen3 Coder· Design Arena Leaderboards data1,146
100GLM 4.5 Air· Design Arena Leaderboards data1,134

Showing 51–100 of 161 results.