Design Arena Leaderboard

312 models across 17 categories · 5.3M+ votes · Updated Jul 25, 2026

Crowdsourced design benchmark from designarena.ai — ELO ratings based on real user preferences

Slides (Agentic)

21 models · 11,900 votes

#	Model	Provider	ELO ▼	Win%	Battles	CI 95%	Avg Time
1	Kimi K3	moonshot	1384	71.3%	303	±0.0	nullms
2	Claude Fable 5	anthropic	1300	63.2%	897	±0.0	6m30s
3	Claude Opus 4.7	anthropic	1276	61.3%	1,366	±0.0	6m36s
4	GPT-5.6 Sol	openai	1252	57.1%	1,124	±0.0	nullms
5	Claude Sonnet 5	anthropic	1244	54.9%	862	±0.0	nullms
6	Claude Opus 4.8	anthropic	1238	55.5%	658	±0.0	7m54s
7	Grok 4.5	xai	1223	51.5%	858	±0.0	nullms
8	Qwen3.7 Max	alibaba	1220	51.6%	393	±0.0	8m57s
9	GLM 5.1	zhipu	1199	51.7%	1,578	±0.0	9m38s
10	Gemini 3.5 Flash	google	1194	55.9%	4,234	±0.0	5m5s
11	GLM 5.2	zhipu	1181	47.4%	709	±0.0	8m33s
12	GPT-5.6 Terra	openai	1178	46.7%	289	±0.0	nullms
13	Muse Spark 1.1	meta	1169	44.1%	1,873	±0.0	nullms
14	GLM 5V Turbo	zhipu	1149	48.3%	713	±0.0	14m15s
15	Kimi K2.7 Code	moonshot	1146	42.3%	638	±0.0	11m49s
16	Kimi K2.6	moonshot	1133	42.0%	1,487	±0.0	10m10s
17	GPT-5.5	openai	1107	41.1%	1,195	±0.0	4m28s
18	Gemini 3.1 Pro Preview	google	1061	32.3%	1,391	±0.0	4m50s
19	Gemini 3 Pro Preview	google	1059	35.3%	691	±0.0	4m59s
20	Grok 4.3	xai	1028	30.8%	786	±0.0	2m33s
21	Gemini 3 Flash Preview	google	1012	38.3%	201	±0.0	4m1s