Design Arena Leaderboard

432 models across 23 categories · 8.0M+ votes · Updated Sep 23, 2026

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

Game Dev (Agentic)

27 models · 20,772 votes

#	Model	Provider	ELO ▼	Win%	Battles	CI 95%	Avg Time
1	Claude Fable 5	anthropic	1285	61.9%	1,428	±9.6	nullms
2	GPT-5.6 Sol (xhigh)	openai	1267	56.7%	1,039	±11.2	nullms
3	Claude Opus 5	anthropic	1261	53.8%	774	±12.7	nullms
4	Kimi K3	moonshot	1252	54.9%	694	±13.3	nullms
5	Claude Opus 4.8	anthropic	1238	55.8%	1,367	±9.6	nullms
6	Gemini 3.7 Flash	google	1235	52.1%	4,703	±5.7	nullms
7	Grok 4.5	xai	1235	55.9%	1,468	±9.4	nullms
8	Claude Sonnet 5	anthropic	1227	52.3%	1,391	±9.5	nullms
9	Claude Opus 4.6	anthropic	1215	50.3%	201	±24.4	nullms
10	Grok 4.6	xai	1210	47.2%	672	±13.7	nullms
11	GLM 5.2	zhipu	1205	49.0%	844	±12.0	nullms
12	GPT-5.6 Sol (Medium)	openai	1199	52.4%	1,151	±10.5	nullms
13	Muse Spark 1.1	meta	1183	48.3%	2,705	±7.2	nullms
14	Gemini 3.5 Flash	google	1180	55.4%	3,580	±6.9	nullms
15	Gemini 3.6 Flash	google	1180	47.8%	1,686	±8.8	nullms
16	GPT-5.5	openai	1179	52.4%	2,274	±7.9	nullms
17	Muse Spark 1.2	meta	1179	42.7%	614	±14.4	nullms
18	Claude Sonnet 4.6	anthropic	1177	52.2%	245	±22.1	nullms
19	GLM 5.1	zhipu	1165	49.8%	1,493	±9.4	nullms
20	Qwen3.7 Max	alibaba	1161	46.6%	699	±13.3	nullms
21	MiniMax M3	minimax	1158	44.6%	504	±15.5	nullms
22	Kimi K2.6	moonshot	1141	47.5%	2,461	±7.7	nullms
23	Kimi K2.7 Code	moonshot	1130	43.0%	832	±12.4	nullms
24	Gemini 3.1 Pro Preview	google	1114	44.2%	1,855	±8.9	nullms
25	GLM 5V Turbo	zhipu	1102	42.6%	1,952	±8.8	nullms
26	GPT-5.6 Terra	openai	1080	35.4%	645	±14.5	nullms
27	Grok 4.3	xai	1007	28.8%	1,908	±9.3	nullms