llm.ing

Design Arena — webapps

mirrorrun by Design Arenaunit: elo42 modelssource ↗

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

websitecodecategoriesdatavizuicomponentgamedev3dsvgasciiartfullstackmobileapps+15 more
#ModelScoreObserved
1Claude Fable 5.1
1337
Sep 21, 2026
2Qwen3.8 Max
1335
Sep 4, 2026
3Muse Spark 1.3
1312
Sep 21, 2026
4Kimi K3
1309
Sep 16, 2026
5Claude Opus 4.7
1279
Sep 1, 2026
5Claude Opus 5
1279
Sep 20, 2026
7Claude Fable 5
1274
Sep 17, 2026
8Kimi K2.6
1268
Jul 23, 2026
9Grok 4.6
1264
Sep 16, 2026
10Claude Sonnet 5
1260
Sep 16, 2026
11Gemini 3.8 Flash
1256
Sep 19, 2026
12Claude Opus 4.8
1246
Sep 18, 2026
13Muse Spark 1.2
1240
Sep 16, 2026
14GLM-5.2
1236
Sep 16, 2026
15Gemini 3.7 Flash
1235
Sep 19, 2026
16Grok 4.5
1231
Sep 16, 2026
17Qwen3.7 Max
1225
Sep 16, 2026
18Claude Opus 4.6
1219
Sep 16, 2026
18Gemini 3.5 Flash
1219
Sep 16, 2026
20Muse Spark 1.1
1217
Sep 16, 2026
21MiniMax-M3
1215
Sep 16, 2026
22Gemini 3.6 Flash
1212
Sep 21, 2026
23Claude Sonnet 4.6
1205
Sep 16, 2026
24Kimi K2.7 Code
1190
Sep 16, 2026
24GLM-5.1
1190
Sep 16, 2026
26Claude Opus 4.5 (latest)
1174
Sep 16, 2026
27Grok 4.20 (Non-Reasoning)
1159
Sep 16, 2026
28Kimi K2.5
1151
Sep 16, 2026
29Grok 4.3
1144
Sep 16, 2026
30Gemini 3.1 Pro Preview
1141
Sep 16, 2026
31GLM-5V-Turbo
1139
Sep 16, 2026
32GPT-5-Codex
1138
Jul 23, 2026
33GPT-5.5
1132
Sep 16, 2026
34Gemini 3 Flash Preview
1126
Sep 16, 2026
35GPT-5.2
1101
Sep 17, 2026
36Claude Sonnet 4.5 (latest)
1075
Sep 16, 2026
37GPT-5.4
1074
Sep 16, 2026
38GPT-5.2 Codex
1066
Sep 19, 2026
39GPT-5.3 Codex
1048
Sep 21, 2026
40GPT-5.1 Codex
1033
Sep 16, 2026
41GPT-5.1
1029
Aug 18, 2026
42DeepSeek V4 Pro
1000
Aug 2, 2026

Dot = elo on a 983–1354 scale (position, not length). Scores link to model pages; every number keeps its source on the model page.