Skip to content

Benchmark

Design Arena Elo (agents/nativeapps)

Agentic benchmarks measure how well models carry out tasks that take several steps. Higher scores are better.

Models measured
42
Scores recorded
43
Sources
1
IQ weight
2.30

Score distribution

1,154.9
Mean
1,168
Median
84.9
Std dev
1,033
10th percentile
1,253
90th percentile
955
Min
9551,401
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.30 for chat models, measured over 42 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Claude Opus 5.5· Design Arena Leaderboards data1,401
2Muse Spark 1.3· Design Arena Leaderboards data1,282
3Claude Fable 5· Design Arena Leaderboards data1,275
4Claude Sonnet 5.5· Design Arena Leaderboards data1,263
5Kimi K3· Design Arena Leaderboards data1,253
6Claude Opus 5· Design Arena Leaderboards data1,243
7Gemini 3.7 Flash· Design Arena Leaderboards data1,231
8Claude Opus 4.8· Design Arena Leaderboards data1,221
9GPT-5.6 Solxhigh· Design Arena Leaderboards data1,215
10Claude Sonnet 5· Design Arena Leaderboards data1,209
11Muse Spark 1.2· Design Arena Leaderboards data1,207
12Gemini 3.6 Flash· Design Arena Leaderboards data1,193
13Qwen3.8 Max· Design Arena Leaderboards data1,191
14GPT-6 Astra· Design Arena Leaderboards data1,190
15GPT-6.1 Sol· Design Arena Leaderboards data1,190
16Claude Sonnet 4.6· Design Arena Leaderboards data1,181
17Claude Opus 4.6· Design Arena Leaderboards data1,179
18GPT-6 Luna· Design Arena Leaderboards data1,174
19Gemini 3.5 Flash· Design Arena Leaderboards data1,173
20GLM 5.1· Design Arena Leaderboards data1,173
21GPT-6 Sol· Design Arena Leaderboards data1,169
22GLM 5.2· Design Arena Leaderboards data1,168
23Kimi K3.1 Pro· Design Arena Leaderboards data1,164
24Claude Opus 4.5· Design Arena Leaderboards data1,161
25MiniMax M3· Design Arena Leaderboards data1,155
26GPT 5.5· Design Arena Leaderboards data1,152
27Kimi K2.6· Design Arena Leaderboards data1,150
28Qwen3.7 Max· Design Arena Leaderboards data1,146
29GPT-5.6 Sol· Design Arena Leaderboards data1,137
30GLM 5· Design Arena Leaderboards data1,134
31Kimi K2.7 Code· Design Arena Leaderboards data1,127
32Kimi K2.5· Design Arena Leaderboards data1,103
33GLM 4.7· Design Arena Leaderboards data1,095
34Gemini 3.8 Flash· Design Arena Leaderboards data1,091
35GLM 4.6· Design Arena Leaderboards data1,070
36GPT-5.6 Terra· Design Arena Leaderboards data1,055
37GPT 5.2· Design Arena Leaderboards data1,044
38Grok 4.20reasoning, 0309· Design Arena Leaderboards data1,035
39Gemini 3 Flash· Design Arena Leaderboards data1,033
40Gemini 3.1 Pro Preview· Design Arena Leaderboards data1,033
41GPT 5.4· Design Arena Leaderboards data1,020
42Gemini 3 Pro Preview· Design Arena Leaderboards data1,018
43Grok 4.3· Design Arena Leaderboards data955