Skip to content

Benchmark

Design Arena Elo (agents/fullstack)

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

Models measured
48
Scores recorded
49
Sources
1
IQ weight
1.58

Score distribution

1,154.4
Mean
1,154
Median
116.3
Std dev
1,004
10th percentile
1,301
90th percentile
958
Min
9581,412
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.58 for chat models, measured over 48 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 5.2· Design Arena Leaderboards data1,412
2Claude Opus 5.5· Design Arena Leaderboards data1,384
3Claude Sonnet 5.5· Design Arena Leaderboards data1,351
4GPT-6 Astra· Design Arena Leaderboards data1,332
5Kimi K3.1 Pro· Design Arena Leaderboards data1,322
6Claude Fable 5.1· Design Arena Leaderboards data1,301
7Kimi K3· Design Arena Leaderboards data1,288
8Claude Opus 5· Design Arena Leaderboards data1,282
9Qwen3.8 Max· Design Arena Leaderboards data1,279
10Muse Spark 1.3· Design Arena Leaderboards data1,270
11Claude Fable 5· Design Arena Leaderboards data1,255
12GPT-5.6 Solxhigh· Design Arena Leaderboards data1,234
13Claude Opus 4.8· Design Arena Leaderboards data1,225
14Claude Sonnet 5· Design Arena Leaderboards data1,224
15Gemini 3.8 Flash· Design Arena Leaderboards data1,217
16GLM 5.2· Design Arena Leaderboards data1,215
17GPT-6 Sol· Design Arena Leaderboards data1,208
18Claude Opus 4.6· Design Arena Leaderboards data1,194
19Claude Sonnet 4.6· Design Arena Leaderboards data1,187
20Muse Spark 1.2· Design Arena Leaderboards data1,184
21Gemini 3.7 Flash· Design Arena Leaderboards data1,179
22MiniMax M3· Design Arena Leaderboards data1,177
23Gemini 3.5 Flash· Design Arena Leaderboards data1,171
24Qwen3.7 Max· Design Arena Leaderboards data1,169
25Gemini 3.6 Flash· Design Arena Leaderboards data1,154
26Kimi K2.7 Code· Design Arena Leaderboards data1,151
27GLM 5.1· Design Arena Leaderboards data1,142
28GPT-6 Luna· Design Arena Leaderboards data1,139
29GPT-5.6 Sol· Design Arena Leaderboards data1,134
30Claude Opus 4.5· Design Arena Leaderboards data1,131
31Kimi K2.6· Design Arena Leaderboards data1,130
32GLM 5V Turbo· Design Arena Leaderboards data1,124
33GLM 5· Design Arena Leaderboards data1,097
34Kimi K2.5· Design Arena Leaderboards data1,088
35GPT 5.5· Design Arena Leaderboards data1,064
36Gemini 3 Pro Preview· Design Arena Leaderboards data1,049
37Gemini 3.1 Pro Preview· Design Arena Leaderboards data1,043
38Gemini 3 Flash· Design Arena Leaderboards data1,038
39GPT-5.6 Terra· Design Arena Leaderboards data1,037
40Grok 4.20reasoning, 0309· Design Arena Leaderboards data1,029
41Claude Sonnet 4.5· Design Arena Leaderboards data1,025
42GLM 4.7· Design Arena Leaderboards data1,024
43GPT 5.2· Design Arena Leaderboards data1,017
44GLM 4.6· Design Arena Leaderboards data1,004
45GPT 5.4· Design Arena Leaderboards data992
46Grok Code Fast 1· Design Arena Leaderboards data984
47Grok 4.3· Design Arena Leaderboards data982
48GPT-5.2 Codex· Design Arena Leaderboards data968
49GPT-5.3 Codex· Design Arena Leaderboards data958