llm.ing

Design Arena — androidnative

mirrorrun by Design Arenaunit: elo36 modelssource ↗

Caveat: Head-to-head human votes on generated designs/code, mirrored via OpenRouter.

websitecodecategoriesdatavizuicomponentgamedev3dsvgasciiartfullstackmobileapps+15 more
#ModelScoreObserved
1Grok 4.6
1302
Sep 21, 2026
2Claude Fable 5
1294
Sep 20, 2026
3GLM-5V-Turbo
1267
Jul 23, 2026
3Claude Opus 5
1267
Sep 21, 2026
5Kimi K3
1259
Sep 21, 2026
6Gemini 3.7 Flash
1253
Sep 20, 2026
7Grok 4.5
1247
Sep 21, 2026
8Claude Opus 4.8
1246
Sep 21, 2026
8Claude Opus 4.7
1246
Sep 6, 2026
10Claude Sonnet 5
1239
Sep 21, 2026
11Muse Spark 1.2
1236
Sep 20, 2026
12Claude Opus 4.6
1216
Sep 19, 2026
13Gemini 3.6 Flash
1213
Sep 20, 2026
14Claude Sonnet 4.6
1210
Sep 20, 2026
15Gemini 3.5 Flash
1197
Sep 20, 2026
16GLM-5.1
1192
Sep 20, 2026
17Claude Opus 4.5 (latest)
1190
Sep 19, 2026
18GLM-5.2
1189
Sep 19, 2026
19Muse Spark 1.1
1186
Sep 21, 2026
20GPT-5.2 Codex
1176
Jul 23, 2026
21GPT-5.5
1175
Sep 20, 2026
22Kimi K2.6
1173
Sep 20, 2026
23Qwen3.7 Max
1169
Sep 20, 2026
24MiniMax-M3
1166
Sep 21, 2026
25GLM-5
1159
Sep 19, 2026
26Kimi K2.7 Code
1150
Sep 20, 2026
27Kimi K2.5
1133
Sep 20, 2026
28GLM-4.7
1128
Sep 20, 2026
29GLM-4.6
1099
Sep 20, 2026
30GPT-5.3 Codex
1084
Jul 23, 2026
31GPT-5.2
1073
Sep 19, 2026
32Gemini 3 Flash Preview
1065
Sep 20, 2026
33Gemini 3.1 Pro Preview
1063
Sep 19, 2026
34Grok 4.20 (Non-Reasoning)
1058
Sep 21, 2026
35GPT-5.4
1045
Sep 19, 2026
36Grok 4.3
972
Sep 19, 2026

Dot = elo on a 955–1319 scale (position, not length). Scores link to model pages; every number keeps its source on the model page.