Skip to content

Benchmark

Design Arena Elo (agents/agon_webapps)

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

Models measured
46
Scores recorded
47
Sources
1
IQ weight
1.62

Score distribution

1,169.2
Mean
1,173
Median
86
Std dev
1,058
10th percentile
1,280
90th percentile
990
Min
9901,312
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.62 for chat models, measured over 46 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 Fable 5.1· Design Arena Leaderboards data1,312
2Claude Opus 5.5· Design Arena Leaderboards data1,311
3GPT-6 Astra· Design Arena Leaderboards data1,308
4Kimi K3.1 Pro· Design Arena Leaderboards data1,298
5Claude Sonnet 5.5· Design Arena Leaderboards data1,291
6Gemini 4 Argon· Design Arena Leaderboards data1,280
7GPT-6.1 Sol· Design Arena Leaderboards data1,274
8Qwen3.8 Max· Design Arena Leaderboards data1,274
9Kimi K3· Design Arena Leaderboards data1,269
10Muse Spark 1.3· Design Arena Leaderboards data1,269
11Claude Opus 5· Design Arena Leaderboards data1,240
12Claude Fable 5· Design Arena Leaderboards data1,234
13Claude Sonnet 5· Design Arena Leaderboards data1,214
14GPT-5.6 Solxhigh· Design Arena Leaderboards data1,214
15Gemini 3.8 Flash· Design Arena Leaderboards data1,208
16Claude Opus 4.8· Design Arena Leaderboards data1,205
17GPT-6 Sol· Design Arena Leaderboards data1,204
18Gemini 3.7 Flash· Design Arena Leaderboards data1,201
19GLM 5.2· Design Arena Leaderboards data1,195
20Muse Spark 1.2· Design Arena Leaderboards data1,195
21MiniMax M3· Design Arena Leaderboards data1,183
22Qwen3.7 Max· Design Arena Leaderboards data1,181
23Claude Opus 4.6· Design Arena Leaderboards data1,178
24Gemini 3.5 Flash· Design Arena Leaderboards data1,173
25Gemini 3.6 Flash· Design Arena Leaderboards data1,170
26Inkling FP4· Design Arena Leaderboards data1,168
27GPT-5.6 Sol· Design Arena Leaderboards data1,163
28Claude Sonnet 4.6· Design Arena Leaderboards data1,162
29GLM 5.1· Design Arena Leaderboards data1,148
30Kimi K2.7 Code· Design Arena Leaderboards data1,148
31GPT-6 Luna· Design Arena Leaderboards data1,145
32Claude Opus 4.5· Design Arena Leaderboards data1,131
33Grok 4.20reasoning, 0309· Design Arena Leaderboards data1,116
34Kimi K2.5· Design Arena Leaderboards data1,108
35Gemini 3.1 Pro Preview· Design Arena Leaderboards data1,098
36GLM 5V Turbo· Design Arena Leaderboards data1,097
37Grok 4.3· Design Arena Leaderboards data1,097
38GPT-5.6 Terra· Design Arena Leaderboards data1,092
39GPT 5.5· Design Arena Leaderboards data1,089
40Gemini 3 Pro Preview· Design Arena Leaderboards data1,086
41Gemini 3 Flash· Design Arena Leaderboards data1,083
42GPT 5.2· Design Arena Leaderboards data1,058
43Claude Sonnet 4.5· Design Arena Leaderboards data1,032
44GPT 5.4· Design Arena Leaderboards data1,031
45GPT-5.2 Codex· Design Arena Leaderboards data1,023
46GPT-5.3 Codex· Design Arena Leaderboards data1,005
47GPT-5.1 Codex· Design Arena Leaderboards data990