Skip to content

Benchmark

Design Arena Elo (models/ascii)

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

Models measured
76
Scores recorded
84
Sources
1
IQ weight
2.49

Score distribution

1,168.1
Mean
1,152
Median
97.1
Std dev
1,048
10th percentile
1,302
90th percentile
946
Min
9461,450
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.49 for chat models, measured over 76 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Gemini 4 Hydrogen· Design Arena Leaderboards data1,450
2Claude Sonnet 5.5· Design Arena Leaderboards data1,399
3Kimi K3· Design Arena Leaderboards data1,352
4GPT-6 Astra· Design Arena Leaderboards data1,344
5Claude Opus 5.5· Design Arena Leaderboards data1,343
6Claude Opus 5· Design Arena Leaderboards data1,338
7Claude Fable 5· Design Arena Leaderboards data1,312
8Claude Fable 5.1· Design Arena Leaderboards data1,309
9Kimi K3.1 Pro· Design Arena Leaderboards data1,302
10GPT-5.6 Solxhigh· Design Arena Leaderboards data1,290
11Muse Spark 1.3· Design Arena Leaderboards data1,284
12Muse Spark 1.2· Design Arena Leaderboards data1,268
13Gemini 3.6 Flash· Design Arena Leaderboards data1,266
14Claude Opus 4.8· Design Arena Leaderboards data1,261
15Gemini 3.8 Flash· Design Arena Leaderboards data1,259
16Gemini 3.7 Flash· Design Arena Leaderboards data1,258
17Gemini 3.1 Pro Preview· Design Arena Leaderboards data1,255
18DeepSeek V4.1 Flash· Design Arena Leaderboards data1,253
19GPT-5.6 Sol· Design Arena Leaderboards data1,252
20Claude Opus 4.6thinking· Design Arena Leaderboards data1,240
21Gemini 3.5 Flash· Design Arena Leaderboards data1,239
22Claude Opus 4.6· Design Arena Leaderboards data1,238
23GPT 5.5· Design Arena Leaderboards data1,237
24DeepSeek V4.1 Flash 1017· Design Arena Leaderboards data1,222
25Claude Sonnet 4.6· Design Arena Leaderboards data1,221
26Claude Haiku 5.5· Design Arena Leaderboards data1,213
27GPT-6 Sol· Design Arena Leaderboards data1,213
28Mistral Large 4· Design Arena Leaderboards data1,196
29Qwen3.7 Max· Design Arena Leaderboards data1,195
30Claude Sonnet 4.5thinking· Design Arena Leaderboards data1,191
31Claude Sonnet 5· Design Arena Leaderboards data1,190
32Qwen3.8 Max Preview· Design Arena Leaderboards data1,181
33Claude Opus 4.5· Design Arena Leaderboards data1,176
34Kimi K2.7 Code· Design Arena Leaderboards data1,175
35GPT 5.4none· Design Arena Leaderboards data1,174
36GLM 5.2· Design Arena Leaderboards data1,173
37GPT 5.4low· Design Arena Leaderboards data1,173
38GPT-5.6 Terra· Design Arena Leaderboards data1,163
39Claude Sonnet 4.5· Design Arena Leaderboards data1,161
40Grok 4.20· Design Arena Leaderboards data1,161
41Claude Opus 4.1· Design Arena Leaderboards data1,153
42Gemini 3.1 Flash Lite· Design Arena Leaderboards data1,152
43GLM 4.7· Design Arena Leaderboards data1,151
44Kimi K2.5· Design Arena Leaderboards data1,151
45GPT-6 Luna· Design Arena Leaderboards data1,150
46GLM 5 Turbo· Design Arena Leaderboards data1,147
47MiMo V2.5 Pro· Design Arena Leaderboards data1,144
48Kimi K2.6· Design Arena Leaderboards data1,143
49MiniMax M3· Design Arena Leaderboards data1,141
50GPT-5.3 Codex· Design Arena Leaderboards data1,137

Showing 1–50 of 84 results.