llm.ing

Design Arena — 3d

mirrorrun by Design Arenaunit: elo68 modelssource ↗

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

websitecodecategoriesdatavizuicomponentgamedev3dsvgasciiartwebappsfullstack+15 more
#ModelScoreObserved
1Kimi K3
1465
Jul 24, 2026
2Claude Fable 5
1366
Jul 24, 2026
3GLM-5.2
1358
Jul 24, 2026
4Gemini 3.6 Flash
1337
Jul 24, 2026
5GLM-5.1
1336
Jul 23, 2026
6Claude Opus 4.6
1335
Jul 24, 2026
7Kimi K2.6
1334
Jul 24, 2026
8Claude Sonnet 5
1321
Jul 23, 2026
9Grok 4.5
1318
Jul 24, 2026
10DeepSeek V4 Pro
1311
Jul 23, 2026
10Qwen3.7 Max
1311
Jul 24, 2026
12Muse Spark 1.1
1308
Jul 24, 2026
12Qwen3.7 Plus
1308
Jul 24, 2026
14Claude Opus 4.7
1305
Jul 23, 2026
15Kimi K2.7 Code
1303
Jul 24, 2026
16Claude Sonnet 4.6
1293
Jul 24, 2026
16Gemini 3.5 Flash
1293
Jul 23, 2026
18Gemini 3.1 Pro Preview
1292
Jul 23, 2026
19GLM-5
1289
Jul 23, 2026
20MiniMax-M3
1281
Jul 24, 2026
21Claude Opus 4.8
1275
Jul 24, 2026
22Claude Opus 4.5 (latest)
1273
Jul 24, 2026
23GLM-5V-Turbo
1270
Jul 23, 2026
24Kimi K2.5
1265
Jul 23, 2026
25Qwen3.6 Plus
1255
Jul 24, 2026
26GPT-5.5
1251
Jul 23, 2026
26Grok 4.20 (Non-Reasoning)
1251
Jul 23, 2026
28GLM-4.7
1250
Jul 24, 2026
29DeepSeek V4 Flash
1249
Jul 24, 2026
30MiniMax-M2.7
1248
Jul 24, 2026
31Gemini 3 Flash Preview
1243
Jul 24, 2026
32GLM-4.5
1230
Jul 24, 2026
33MiniMax-M2.5
1229
Jul 24, 2026
34MiniMax-M2.1
1225
Jul 23, 2026
35Qwen3.5 397B-A17B
1220
Jul 24, 2026
36Claude Sonnet 4.5 (latest)
1219
Jul 23, 2026
37Claude Opus 4.1 (latest)
1211
Jul 23, 2026
38DeepSeek V3.2
1193
Jul 24, 2026
39Nemotron 3 Ultra 550B A55B
1192
Jul 23, 2026
40GLM-4.6
1187
Jul 24, 2026
41Grok 4.3
1183
Jul 24, 2026
42GLM-4.5-Air
1182
Jul 24, 2026
43GLM-4.7-Flash
1180
Jul 24, 2026
44Qwen3.5 Plus
1179
Jul 23, 2026
44Step 3.7 Flash
1179
Jul 24, 2026
46DeepSeek R1
1171
Jul 24, 2026
47GPT-5.4
1161
Jul 23, 2026
48MiniMax-M2
1157
Jul 24, 2026
49DeepSeek Chat
1147
Jul 23, 2026
50Gemini 2.5 Pro
1142
Jul 24, 2026
51GPT-5.2
1139
Jul 24, 2026
52Qwen3 Max
1133
Jul 23, 2026
53Claude Haiku 4.5 (latest)
1132
Jul 23, 2026
54Gemini 2.5 Flash
1131
Jul 23, 2026
55GPT-5.1
1121
Jul 24, 2026
56GPT-5
1116
Jul 23, 2026
57Gemini 3.1 Flash Lite Preview
1108
Jul 23, 2026
58GPT-5 Mini
1097
Jul 24, 2026
59GPT-5.3 Codex
1068
Jul 23, 2026
60Qwen3 235B-A22Bthinking
1059
Jul 23, 2026
61GPT-5.1 Codex mini
1049
Jul 23, 2026
62Mercury 2
1047
Jul 24, 2026
63GPT-5 Nano
1024
Jul 23, 2026
64GPT-4.1 nano
984
Jul 24, 2026
65GPT-4o
928
Jul 24, 2026
66o4-mini
913
Jul 23, 2026
67GPT-4.1
909
Jul 24, 2026
68GPT-4.1 mini
896
Jul 24, 2026

Dot = elo on a 867–1494 scale (position, not length). Scores link to model pages; every number keeps its source on the model page.