llm.ing

Design Arena — codecategories

mirrorrun by Design Arenaunit: elo71 modelssource ↗

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

websitecodecategoriesdatavizuicomponentgamedev3dsvgasciiartwebappsfullstack+15 more
#ModelScoreObserved
1Kimi K3
1431
Jul 24, 2026
2GLM-5.2
1341
Jul 23, 2026
3Claude Fable 5
1339
Jul 23, 2026
4Claude Opus 4.6
1326
Jul 23, 2026
5Gemini 3.6 Flash
1322
Jul 23, 2026
6Claude Opus 4.7
1320
Jul 24, 2026
7Grok 4.5
1314
Jul 24, 2026
8Claude Sonnet 5
1312
Jul 23, 2026
9Kimi K2.6
1310
Jul 24, 2026
9GLM-5.1
1310
Jul 24, 2026
11Claude Sonnet 4.6
1308
Jul 24, 2026
12Muse Spark 1.1
1303
Jul 24, 2026
13Qwen3.7 Max
1299
Jul 23, 2026
14Kimi K2.7 Code
1291
Jul 23, 2026
15Qwen3.7 Plus
1290
Jul 23, 2026
16Gemini 3.5 Flash
1289
Jul 23, 2026
17MiniMax-M3
1287
Jul 23, 2026
18GPT-5.5
1284
Jul 24, 2026
19GLM-5
1279
Jul 23, 2026
20Claude Opus 4.5 (latest)
1275
Jul 24, 2026
21Gemini 3.1 Pro Preview
1274
Jul 23, 2026
22DeepSeek V4 Pro
1270
Jul 24, 2026
22Claude Opus 4.8
1270
Jul 23, 2026
22Kimi K2.5
1270
Jul 23, 2026
25MiniMax-M2.7
1266
Jul 23, 2026
26GLM-5V-Turbo
1261
Jul 23, 2026
27Qwen3.6 Plus
1256
Jul 24, 2026
28Grok 4.20 (Non-Reasoning)
1251
Jul 23, 2026
29GLM-4.7
1249
Jul 23, 2026
30GPT-5.4
1244
Jul 24, 2026
31MiniMax-M2.5
1240
Jul 23, 2026
32DeepSeek V4 Flash
1238
Jul 24, 2026
33MiniMax-M2.1
1223
Jul 24, 2026
34Gemini 3 Flash Preview
1222
Jul 24, 2026
35Grok 4.3
1221
Jul 23, 2026
36Claude Sonnet 4.5 (latest)
1215
Jul 24, 2026
37Qwen3.5 397B-A17B
1212
Jul 24, 2026
38GLM-4.7-Flash
1210
Jul 24, 2026
39Claude Opus 4.1 (latest)
1204
Jul 23, 2026
39GPT-5.1
1204
Jul 23, 2026
41GPT-5.2
1202
Jul 24, 2026
42GPT-5
1201
Jul 24, 2026
43Qwen3.5 Plus
1200
Jul 23, 2026
43Step 3.7 Flash
1200
Jul 24, 2026
45GLM-4.6
1199
Jul 23, 2026
46GLM-4.5
1198
Jul 23, 2026
47DeepSeek V3.2
1196
Jul 24, 2026
48Gemini 2.5 Pro
1187
Jul 23, 2026
49GPT-5.1 Codex
1183
Jul 24, 2026
50GPT-5.3 Codex
1177
Jul 24, 2026
51DeepSeek R1
1171
Jul 24, 2026
52GLM-4.5-Air
1170
Jul 24, 2026
53MiniMax-M2
1168
Jul 23, 2026
54Nemotron 3 Ultra 550B A55B
1157
Jul 23, 2026
55Claude Haiku 4.5 (latest)
1147
Jul 24, 2026
55GPT-5 Mini
1147
Jul 24, 2026
57Qwen3 Max
1142
Jul 24, 2026
58DeepSeek Chat
1141
Jul 24, 2026
59Gemini 2.5 Flash
1137
Jul 23, 2026
60Kimi K2 Thinking
1131
Jul 23, 2026
61GPT-5.1 Codex mini
1126
Jul 24, 2026
62GPT-5 Nano
1117
Jul 24, 2026
63Gemini 3.1 Flash Lite Preview
1104
Jul 24, 2026
64Qwen3 235B-A22B
1071
Jul 24, 2026
65GPT-4.1
1058
Jul 24, 2026
66o3
1053
Jul 23, 2026
67GPT-4.1 mini
1026
Jul 24, 2026
67Mercury 2
1026
Jul 23, 2026
69o4-mini
1009
Jul 23, 2026
70GPT-4.1 nano
996
Jul 23, 2026
71GPT-4o
893
Jul 24, 2026

Dot = elo on a 866–1458 scale (position, not length). Scores link to model pages; every number keeps its source on the model page.