Skip to content

Benchmark

Design Arena Elo (models/gamedev)

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

Models measured
148
Scores recorded
160
Sources
1
IQ weight
1.91

Score distribution

1,136.4
Mean
1,134
Median
150.8
Std dev
942
10th percentile
1,317
90th percentile
745
Min
7451,540
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.91 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
51DeepSeek V4 Pro· Design Arena Leaderboards data1,204
52Hy4 Flash· Design Arena Leaderboards data1,203
53Claude Opus 4.5· Design Arena Leaderboards data1,201
54GPT-5.6 Terra· Design Arena Leaderboards data1,201
55Kimi K2.7 Code· Design Arena Leaderboards data1,193
56Mistral Large 4· Design Arena Leaderboards data1,190
57GLM 5V Turbo· Design Arena Leaderboards data1,189
58ERNIE 5.5 Flash Preview· Design Arena Leaderboards data1,187
59Mercury 3· Design Arena Leaderboards data1,187
60Qwen3.6 Plus· Design Arena Leaderboards data1,184
61MiniMax M3· Design Arena Leaderboards data1,182
62DeepSeek V4 Flash 0731· Design Arena Leaderboards data1,181
63Kimi K2.5· Design Arena Leaderboards data1,178
64Gemini 3.1 Pro Preview· Design Arena Leaderboards data1,174
65GPT-5.6 Luna· Design Arena Leaderboards data1,172
66DeepSeek V4 Flash· Design Arena Leaderboards data1,169
67GPT 5.4none· Design Arena Leaderboards data1,166
68Grok 4.20· Design Arena Leaderboards data1,164
69GPT 5.4low· Design Arena Leaderboards data1,163
70GLM 4.7· Design Arena Leaderboards data1,160
71GPT 5high· Design Arena Leaderboards data1,157
72GPT 5.1high· Design Arena Leaderboards data1,150
73Claude Opus 4.1thinking, 20250805· Design Arena Leaderboards data1,143
74Grok 4.3· Design Arena Leaderboards data1,143
75Claude Opus 4· Design Arena Leaderboards data1,142
76Claude Sonnet 4.5· Design Arena Leaderboards data1,138
77Gemini 3 Flash· Design Arena Leaderboards data1,138
78Claude Opus 4.1· Design Arena Leaderboards data1,135
79Claude Sonnet 4.5thinking· Design Arena Leaderboards data1,135
80Inkling FP4· Design Arena Leaderboards data1,134
81GPT-5.3 Codex· Design Arena Leaderboards data1,133
82GPT 5· Design Arena Leaderboards data1,127
83Step 3.7 Flash· Design Arena Leaderboards data1,124
84GPT 5.1low· Design Arena Leaderboards data1,123
85GLM 4.6· Design Arena Leaderboards data1,122
86GLM 4.5· Design Arena Leaderboards data1,117
87DeepSeek V3.2 Exp· Design Arena Leaderboards data1,116
88Qwen3.5 397B A17B· Design Arena Leaderboards data1,112
89Claude Sonnet 4thinking· Design Arena Leaderboards data1,111
90Claude Sonnet 3.7· Design Arena Leaderboards data1,110
91GPT-5.1 Codex· Design Arena Leaderboards data1,110
92Hy3 (Deep Infra)· Design Arena Leaderboards data1,108
93Nemotron 3 Ultra· Design Arena Leaderboards data1,108
94Claude Sonnet 4· Design Arena Leaderboards data1,106
95GLM 4.7 Flash· Design Arena Leaderboards data1,106
96MiniMax M2.1· Design Arena Leaderboards data1,106
97GPT 5.1none· Design Arena Leaderboards data1,104
98DeepSeek V3.1thinking· Design Arena Leaderboards data1,103
99DeepSeek V3.2· Design Arena Leaderboards data1,103
100MiMo V2 Flash· Design Arena Leaderboards data1,102

Showing 51–100 of 160 results.