llm.ing

Design Arena — website

mirrorrun by Design Arenaunit: elo72 modelssource ↗

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

websitecodecategoriesdatavizuicomponentgamedev3dsvgasciiartwebappsfullstack+15 more
#ModelScoreObserved
1Kimi K3
1371
Jul 24, 2026
2GLM-5.2
1338
Jul 23, 2026
3Claude Fable 5
1328
Jul 23, 2026
4Gemini 3.6 Flash
1323
Jul 24, 2026
5Claude Opus 4.6
1322
Jul 23, 2026
5Claude Opus 4.7
1322
Jul 23, 2026
7Grok 4.5
1321
Jul 23, 2026
8Claude Sonnet 5
1314
Jul 23, 2026
9Claude Sonnet 4.6
1312
Jul 23, 2026
10Kimi K2.6
1304
Jul 23, 2026
11Kimi K2.7 Code
1301
Jul 23, 2026
11GLM-5.1
1301
Jul 24, 2026
13Qwen3.7 Plus
1296
Jul 24, 2026
14Muse Spark 1.1
1293
Jul 24, 2026
15Qwen3.7 Max
1292
Jul 24, 2026
16MiniMax-M3
1287
Jul 24, 2026
17Gemini 3.5 Flash
1284
Jul 23, 2026
18GPT-5.5
1281
Jul 23, 2026
19Gemini 3.1 Pro Preview
1278
Jul 23, 2026
20Kimi K2.5
1277
Jul 23, 2026
21GLM-5
1275
Jul 23, 2026
22Claude Opus 4.5 (latest)
1274
Jul 23, 2026
23MiniMax-M2.7
1273
Jul 23, 2026
24Claude Opus 4.8
1269
Jul 23, 2026
25DeepSeek V4 Pro
1261
Jul 23, 2026
26GLM-5V-Turbo
1256
Jul 23, 2026
27Grok 4.20 (Non-Reasoning)
1254
Jul 23, 2026
28GLM-4.7
1253
Jul 23, 2026
29MiniMax-M2.5
1249
Jul 23, 2026
30GPT-5.4
1247
Jul 23, 2026
30Qwen3.6 Plus
1247
Jul 23, 2026
32DeepSeek V4 Flash
1235
Jul 23, 2026
33MiniMax-M2.1
1229
Jul 24, 2026
34Gemini 3 Flash Preview
1223
Jul 23, 2026
35GPT-5.2
1222
Jul 23, 2026
36Grok 4.3
1221
Jul 23, 2026
36GLM-4.7-Flash
1221
Jul 23, 2026
38Qwen3.5 397B-A17B
1218
Jul 23, 2026
39Claude Sonnet 4.5 (latest)
1216
Jul 23, 2026
40GPT-5.1
1215
Jul 23, 2026
40Qwen3.5 Plus
1215
Jul 23, 2026
42GPT-5
1212
Jul 24, 2026
43Step 3.7 Flash
1209
Jul 23, 2026
44Claude Opus 4.1 (latest)
1204
Jul 23, 2026
45GLM-4.6
1202
Jul 23, 2026
46DeepSeek V3.2
1201
Jul 23, 2026
47GLM-4.5
1197
Jul 23, 2026
48Gemini 2.5 Pro
1194
Jul 23, 2026
49GPT-5.3 Codex
1190
Jul 23, 2026
50GPT-5.1 Codex
1188
Jul 23, 2026
51DeepSeek R1
1177
Jul 24, 2026
52GLM-4.5-Air
1174
Jul 23, 2026
53MiniMax-M2
1170
Jul 23, 2026
54GPT-5 Mini
1152
Jul 23, 2026
55Claude Haiku 4.5 (latest)
1149
Jul 23, 2026
56DeepSeek Chat
1147
Jul 23, 2026
57Qwen3 Max
1146
Jul 23, 2026
58Gemini 2.5 Flash
1143
Jul 24, 2026
59Kimi K2 Thinkingthinking
1140
Jul 23, 2026
60GPT-5.1 Codex mini
1138
Jul 23, 2026
61GPT-5 Nano
1129
Jul 23, 2026
61Nemotron 3 Ultra 550B A55B
1129
Jul 24, 2026
63Qwen3-Coder 30B-A3B Instruct
1115
Jul 23, 2026
64Gemini 3.1 Flash Lite Preview
1109
Jul 23, 2026
65Qwen3 235B-A22B
1085
Jul 23, 2026
66GPT-4.1
1066
Jul 23, 2026
67o3
1064
Jul 24, 2026
68GPT-4.1 mini
1025
Jul 23, 2026
69Mercury 2
1016
Jul 23, 2026
70o4-mini
1013
Jul 24, 2026
71GPT-4.1 nano
1000
Jul 23, 2026
72GPT-4o
859
Jul 24, 2026

Dot = elo on a 833–1397 scale (position, not length). Scores link to model pages; every number keeps its source on the model page.