llm.ing

Design Arena — gamedev

mirrorrun by Design Arenaunit: elo81 modelssource ↗

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

websitecodecategoriesdatavizuicomponentgamedev3dsvgasciiartfullstackmobileapps+15 more
#ModelScoreObserved
1Kimi K3
1407
Sep 20, 2026
2Claude Fable 5.1
1403
Sep 21, 2026
3GLM-5.3
1372
Sep 21, 2026
4Muse Spark 1.3
1371
Sep 20, 2026
5Claude Opus 5
1362
Sep 21, 2026
6Claude Fable 5
1358
Sep 21, 2026
7Gemini 3.7 Flash
1350
Sep 21, 2026
8Gemini 3.8 Flash
1341
Sep 21, 2026
9Qwen3.8 Max
1331
Sep 4, 2026
10Muse Spark 1.2
1323
Sep 20, 2026
11Claude Opus 4.7
1321
Sep 6, 2026
11GPT-5.5
1321
Sep 21, 2026
11Grok 4.6
1321
Sep 19, 2026
14Claude Sonnet 5
1314
Sep 20, 2026
15GLM-5.3-Flash
1310
Sep 20, 2026
16Claude Opus 4.6
1301
Sep 19, 2026
17Muse Spark 1.1
1300
Sep 21, 2026
18GLM-5.2
1296
Sep 19, 2026
19Qwen3.7 Max
1292
Sep 20, 2026
20Grok 4.5
1289
Sep 20, 2026
21Qwen3.7 Plus
1288
Sep 21, 2026
22Gemini 3.5 Flash
1287
Sep 20, 2026
23Gemini 3.6 Flash
1286
Sep 20, 2026
23GLM-5.1
1286
Sep 21, 2026
25Claude Sonnet 4.6
1283
Sep 18, 2026
26Claude Opus 4.8
1281
Sep 20, 2026
27Kimi K2.6
1275
Sep 18, 2026
28GPT-5.4
1262
Sep 18, 2026
29GLM-5
1258
Sep 18, 2026
30DeepSeek V4 Pro
1252
Sep 11, 2026
30Claude Opus 4.5 (latest)
1252
Sep 19, 2026
32Kimi K2.7 Code
1244
Sep 20, 2026
33GLM-5V-Turbo
1242
Sep 20, 2026
34MiniMax-M3
1240
Sep 20, 2026
35Qwen3.6 Plus
1234
Sep 21, 2026
36DeepSeek V4 Flash
1233
Sep 18, 2026
37MiniMax-M2.7
1231
Sep 18, 2026
38Kimi K2.5
1230
Sep 18, 2026
39Gemini 3.1 Pro Preview
1226
Sep 18, 2026
40GPT-5.2
1219
Sep 18, 2026
41Grok 4.20 (Non-Reasoning)
1217
Sep 20, 2026
42GLM-4.7
1212
Sep 18, 2026
43GPT-5
1209
Sep 18, 2026
44GPT-5.1
1202
Sep 18, 2026
45MiniMax-M2.5
1200
Sep 18, 2026
46Grok 4.3
1198
Sep 18, 2026
47Claude Opus 4.1 (latest)
1195
Sep 18, 2026
48Claude Sonnet 4.5 (latest)
1190
Sep 18, 2026
48Gemini 3 Flash Preview
1190
Sep 21, 2026
50GPT-5.3 Codex
1185
Sep 18, 2026
51Step 3.7 Flash
1180
Sep 21, 2026
52GLM-4.6
1174
Sep 18, 2026
53GLM-4.5
1168
Sep 11, 2026
54Qwen3.5 397B-A17B
1164
Sep 20, 2026
55GPT-5.1 Codex
1162
Sep 21, 2026
56GLM-4.7-Flash
1158
Sep 11, 2026
56MiniMax-M2.1
1158
Sep 18, 2026
58Nemotron 3 Ultra 550B A55B
1156
Sep 20, 2026
59DeepSeek V3.2
1155
Sep 18, 2026
60GPT-5 Mini
1153
Sep 18, 2026
61MiniMax-M2
1141
Sep 11, 2026
62Gemini 2.5 Pro
1135
Sep 18, 2026
63Qwen3.5 Plus
1129
Sep 18, 2026
64DeepSeek R1
1121
Sep 18, 2026
64Claude Haiku 4.5 (latest)
1121
Sep 21, 2026
66Qwen3 Max
1118
Sep 18, 2026
66GLM-4.5-Air
1118
Sep 20, 2026
68GPT-5.1 Codex mini
1116
Sep 18, 2026
69GPT-4.1
1103
Sep 18, 2026
70GPT-4.1 mini
1094
Sep 18, 2026
71Gemini 2.5 Flash
1089
Sep 18, 2026
72DeepSeek Chat
1077
Sep 18, 2026
73GPT-5 Nano
1068
Sep 18, 2026
74o3
1058
Sep 18, 2026
75Gemini 3.1 Flash Lite Preview
1053
Sep 18, 2026
76o4-mini
1026
Sep 11, 2026
77Mercury 2
1009
Sep 21, 2026
78Kimi K2 Thinking
997
Sep 11, 2026
79GPT-4.1 nano
994
Sep 18, 2026
80Qwen3 235B-A22Bthinking
980
Sep 19, 2026
81GPT-4o
930
Sep 18, 2026

Dot = elo on a 906–1431 scale (position, not length). Scores link to model pages; every number keeps its source on the model page.