llm.ing

Design Arena — website

mirrorrun by Design Arenaunit: elo82 modelssource ↗

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

websitecodecategoriesdatavizuicomponentgamedev3dsvgasciiartfullstackmobileapps+15 more
#ModelScoreObserved
1Muse Spark 1.3
1365
Sep 21, 2026
2Kimi K3
1352
Sep 20, 2026
3Claude Fable 5.1
1322
Sep 21, 2026
4Claude Opus 5
1321
Sep 19, 2026
4Muse Spark 1.2
1321
Sep 17, 2026
6Gemini 3.7 Flash
1316
Sep 21, 2026
7GLM-5.3
1314
Sep 20, 2026
7Gemini 3.8 Flash
1314
Sep 20, 2026
9Gemini 3.6 Flash
1311
Sep 21, 2026
10Claude Fable 5
1309
Sep 21, 2026
11Claude Opus 4.7
1307
Sep 5, 2026
12GLM-5.2
1304
Sep 19, 2026
13Claude Opus 4.6
1303
Sep 17, 2026
13Grok 4.6
1303
Sep 21, 2026
15Claude Sonnet 4.6
1297
Sep 11, 2026
16Grok 4.5
1296
Sep 11, 2026
17Qwen3.8 Max
1295
Sep 4, 2026
18GLM-5.1
1290
Aug 14, 2026
19Claude Sonnet 5
1288
Sep 20, 2026
20Qwen3.7 Max
1285
Sep 14, 2026
21GLM-5.3-Flash
1284
Sep 19, 2026
22Qwen3.7 Plus
1283
Sep 20, 2026
23Muse Spark 1.1
1282
Sep 18, 2026
23Kimi K2.6
1282
Sep 11, 2026
25Kimi K2.7 Code
1279
Sep 18, 2026
26Gemini 3.5 Flash
1276
Sep 20, 2026
27MiniMax-M3
1272
Sep 19, 2026
28GPT-5.5
1269
Sep 21, 2026
29Claude Opus 4.8
1267
Sep 11, 2026
30Gemini 3.1 Pro Preview
1265
Sep 11, 2026
31Kimi K2.5
1261
Sep 11, 2026
32GLM-5
1260
Sep 11, 2026
33Claude Opus 4.5 (latest)
1259
Sep 11, 2026
34MiniMax-M2.7
1257
Sep 20, 2026
35Qwen3.6 Plus
1253
Sep 19, 2026
36DeepSeek V4 Flash
1251
Sep 16, 2026
37DeepSeek V4 Pro
1247
Sep 11, 2026
38GLM-5V-Turbo
1243
Sep 21, 2026
39Grok 4.20 (Non-Reasoning)
1242
Sep 9, 2026
40GLM-4.7
1238
Sep 11, 2026
41MiniMax-M2.5
1235
Sep 20, 2026
42GPT-5.4
1232
Sep 11, 2026
43MiniMax-M2.1
1214
Sep 15, 2026
44Gemini 3 Flash Preview
1208
Sep 11, 2026
45GPT-5.2
1207
Sep 11, 2026
45Grok 4.3
1207
Sep 17, 2026
45Step 3.7 Flash
1207
Sep 20, 2026
48GLM-4.7-Flash
1206
Sep 17, 2026
49Qwen3.5 397B-A17B
1203
Sep 11, 2026
50Claude Sonnet 4.5 (latest)
1202
Sep 11, 2026
51Qwen3.5 Plus
1200
Sep 11, 2026
52GPT-5.1
1199
Sep 11, 2026
53GPT-5
1197
Sep 10, 2026
54Claude Opus 4.1 (latest)
1189
Sep 11, 2026
55DeepSeek V3.2
1187
Sep 11, 2026
55GLM-4.6
1187
Sep 20, 2026
57GLM-4.5
1182
Sep 11, 2026
58Gemini 2.5 Pro
1179
Sep 11, 2026
59GPT-5.3 Codex
1175
Sep 11, 2026
60GPT-5.1 Codex
1174
Sep 11, 2026
61DeepSeek R1
1161
Sep 17, 2026
62GLM-4.5-Air
1159
Sep 11, 2026
63MiniMax-M2
1155
Sep 11, 2026
64Nemotron 3 Ultra 550B A55B
1149
Sep 20, 2026
65GPT-5 Mini
1137
Sep 17, 2026
66Claude Haiku 4.5 (latest)
1135
Sep 19, 2026
67DeepSeek Chat
1132
Sep 11, 2026
68Qwen3 Max
1131
Sep 11, 2026
69Gemini 2.5 Flash
1127
Sep 11, 2026
70Kimi K2 Thinkingthinking
1125
Sep 11, 2026
71GPT-5.1 Codex mini
1123
Sep 11, 2026
72GPT-5 Nano
1114
Sep 19, 2026
73Qwen3-Coder 30B-A3B Instruct
1100
Sep 15, 2026
74Gemini 3.1 Flash Lite Preview
1095
Sep 10, 2026
75Qwen3 235B-A22B
1070
Sep 11, 2026
76GPT-4.1
1051
Sep 11, 2026
77o3
1048
Sep 17, 2026
78Mercury 2
1038
Sep 19, 2026
79GPT-4.1 mini
1010
Sep 14, 2026
80o4-mini
998
Sep 20, 2026
81GPT-4.1 nano
985
Sep 11, 2026
82GPT-4o
843
Sep 17, 2026

Dot = elo on a 816–1392 scale (position, not length). Scores link to model pages; every number keeps its source on the model page.