llm.ing

Design Arena — dataviz

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
1373
Jul 24, 2026
2GLM-5.1
1366
Jul 23, 2026
3Claude Fable 5
1351
Jul 23, 2026
4Gemini 3.6 Flash
1346
Jul 24, 2026
5Muse Spark 1.1
1323
Jul 24, 2026
6Claude Opus 4.6
1317
Jul 24, 2026
7Claude Sonnet 4.6
1315
Jul 24, 2026
8Claude Opus 4.7
1312
Jul 23, 2026
9Qwen3.7 Max
1302
Jul 24, 2026
10Grok 4.5
1301
Jul 23, 2026
11GLM-5.2
1300
Jul 24, 2026
12Kimi K2.6
1292
Jul 24, 2026
13GPT-5.5
1289
Jul 24, 2026
14Claude Sonnet 5
1281
Jul 23, 2026
15Gemini 2.5 Pro
1280
Jul 24, 2026
16Claude Opus 4.5 (latest)
1276
Jul 24, 2026
17MiniMax-M3
1275
Jul 24, 2026
18GPT-5
1272
Jul 24, 2026
19Claude Opus 4.8
1268
Jul 24, 2026
20GPT-5.4
1266
Jul 24, 2026
20MiniMax-M2.7
1266
Jul 23, 2026
22Kimi K2.7 Code
1263
Jul 24, 2026
23GLM-5
1261
Jul 24, 2026
24Gemini 3.1 Pro Preview
1260
Jul 24, 2026
25Kimi K2.5
1258
Jul 24, 2026
26Gemini 3.5 Flash
1257
Jul 24, 2026
27Qwen3.7 Plus
1255
Jul 23, 2026
28Qwen3.6 Plus
1252
Jul 24, 2026
29MiniMax-M2.1
1240
Jul 24, 2026
30Grok 4.20 (Non-Reasoning)
1239
Jul 23, 2026
31GPT-5.1
1238
Jul 24, 2026
32GPT-5.2
1235
Jul 24, 2026
33GLM-5V-Turbo
1231
Jul 24, 2026
34GLM-4.7
1228
Jul 24, 2026
35GLM-4.5-Air
1227
Jul 23, 2026
36Grok 4.3
1219
Jul 24, 2026
37DeepSeek R1
1212
Jul 24, 2026
38DeepSeek V4 Pro
1210
Jul 24, 2026
39Qwen3.5 397B-A17B
1209
Jul 24, 2026
40GPT-5.1 Codex
1206
Jul 24, 2026
40MiniMax-M2.5
1206
Jul 23, 2026
42Claude Sonnet 4.5 (latest)
1202
Jul 24, 2026
43Claude Opus 4.1 (latest)
1200
Jul 24, 2026
43o3
1200
Jul 23, 2026
45GLM-4.6
1198
Jul 24, 2026
46Step 3.7 Flash
1197
Jul 24, 2026
47GPT-5.3 Codex
1196
Jul 24, 2026
48GLM-4.5
1194
Jul 24, 2026
49DeepSeek V3.2
1193
Jul 24, 2026
50MiniMax-M2
1174
Jul 24, 2026
51Qwen3.5 Plus
1168
Jul 24, 2026
52GPT-5 Mini
1162
Jul 24, 2026
53Gemini 2.5 Flash
1161
Jul 24, 2026
54DeepSeek V4 Flash
1158
Jul 23, 2026
54Claude Haiku 4.5 (latest)
1158
Jul 23, 2026
56GLM-4.7-Flash
1157
Jul 24, 2026
57Nemotron 3 Ultra 550B A55B
1147
Jul 24, 2026
58Qwen3 Max
1139
Jul 23, 2026
59GPT-4.1
1138
Jul 24, 2026
60DeepSeek Chat
1132
Jul 24, 2026
61GPT-5.1 Codex mini
1131
Jul 23, 2026
62Qwen3-Coder 30B-A3B Instruct
1118
Jul 24, 2026
63GPT-5 Nano
1094
Jul 24, 2026
64Qwen3 235B-A22B
1092
Jul 24, 2026
65Gemini 3.1 Flash Lite Preview
1078
Jul 24, 2026
66GPT-4.1 mini
1068
Jul 23, 2026
67Kimi K2 Thinking
1053
Jul 24, 2026
68o4-mini
1025
Jul 24, 2026
69Mercury 2
1008
Jul 24, 2026
70GPT-4.1 nano
926
Jul 24, 2026
71GPT-4o
891
Jul 24, 2026

Dot = elo on a 866–1398 scale (position, not length). Scores link to model pages; every number keeps its source on the model page.