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

Mobile (Kotlin)

29 models · 15,555 votes

#	Model	Provider	ELO ▼	Win%	Battles	CI 95%	Avg Time
1	Claude Opus 4.8	anthropic	1335	66.5%	555	±0.0	nullms
2	Claude Fable 5	anthropic	1323	64.5%	394	±0.0	nullms
3	Grok 4.5	xai	1285	59.4%	611	±0.0	nullms
4	Kimi K2.6	moonshot	1283	61.8%	1,351	±0.0	nullms
5	Kimi K2.7 Code	moonshot	1264	58.0%	664	±0.0	nullms
6	Claude Opus 4.7	anthropic	1259	56.0%	259	±0.0	nullms
7	Claude Sonnet 5	anthropic	1244	52.8%	318	±0.0	nullms
8	GPT-5.5	openai	1236	54.2%	1,590	±0.0	nullms
9	GLM 5.1	zhipu	1234	53.1%	1,034	±0.0	nullms
10	GLM 5.2	zhipu	1234	54.7%	2,037	±0.0	nullms
11	GLM 5	zhipu	1222	59.5%	1,999	±0.0	nullms
12	Gemini 3.5 Flash	google	1217	51.6%	2,828	±0.0	nullms
13	GPT-5.6 Sol	openai	1210	48.8%	420	±0.0	nullms
14	Claude Sonnet 4.6	anthropic	1206	61.4%	1,113	±0.0	nullms
15	Claude Opus 4.6	anthropic	1191	68.8%	558	±0.0	nullms
16	Qwen3.7 Max	alibaba	1182	45.6%	781	±0.0	nullms
17	Claude Opus 4.5	anthropic	1164	65.5%	661	±0.0	nullms
18	GPT-5.6 Terra	openai	1155	40.3%	524	±0.0	nullms
19	Grok 4.20 Reasoning	xai	1152	41.5%	803	±0.0	nullms
20	MiniMax M3	minimax	1110	33.7%	635	±0.0	nullms
21	Kimi K2.5 (Thinking)	moonshot	1107	57.9%	551	±0.0	nullms
22	GLM 4.7	zhipu	1101	56.0%	420	±0.0	nullms
23	GLM 4.6	zhipu	1072	52.6%	498	±0.0	nullms
24	Grok 4.3	xai	1054	27.9%	1,634	±0.0	nullms
25	GPT 5.2	openai	1047	49.2%	659	±0.0	nullms
26	Gemini 3 Flash Preview	google	1037	48.0%	719	±0.0	nullms
27	Gemini 3.1 Pro Preview	google	1033	38.6%	5,745	±0.0	nullms
28	Gemini 3 Pro Preview	google	1022	46.2%	701	±0.0	nullms
29	GPT-5.4	openai	1015	47.4%	502	±0.0	nullms