llm.ing

Design Arena — agentichtmlslides

mirrorrun by Design Arenaunit: eloobserved Jul 23, 202610 modelssource ↗

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

websitecodecategoriesdatavizuicomponentgamedev3dsvgasciiartwebappsfullstack+15 more
#ModelScore
1Claude Fable 5
1254
2Kimi K2.6
1248
3Claude Opus 4.7
1243
4Claude Opus 4.8
1227
5Gemini 3.1 Pro Preview
1226
6GLM-5.1
1205
7Gemini 3.5 Flash
1162
8GLM-5V-Turbo
1134
9GPT-5.5
1084
10Grok 4.3
1066

Dot = elo on a 1056–1264 scale (position, not length). Scores link to model pages; every number keeps its source on the model page.