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

ASCII

88 models · 19,727 votes

#	Model	Provider	ELO ▼	Win%	Battles	CI 95%	Avg Time
1	Claude Opus 5	anthropic	1372	69.3%	140	±32.5	4m7s
2	GPT-6 Astra (xhigh)	openai	1354	64.9%	114	±37.1	5m21s
3	Claude Fable 5.1	anthropic	1340	65.3%	121	±34.5	2m58s
4	Claude Fable 5	anthropic	1339	67.8%	283	±22.6	1m27s
5	Muse Spark 1.2	meta	1310	57.9%	368	±19.6	3m36s
6	GPT-5.6 Sol (xhigh)	openai	1305	59.1%	137	±31.3	1m50s
7	Muse Spark 1.1	meta	1298	57.7%	593	±15.5	2m54s
8	Gemini 3.6 Flash	google	1296	56.7%	268	±22.5	1m2s
9	Grok 4.6	xai	1292	58.6%	454	±17.5	2m43s
10	Claude Opus 4.8	anthropic	1291	62.5%	376	±18.8	28.5s
11	Gemini 3.1 Pro Preview	google	1285	63.5%	1,357	±10.4	3m45s
12	GLM-5.3-Flash	zhipu	1282	55.1%	354	±19.8	7m9s
13	GPT-5.5	openai	1275	60.5%	468	±16.8	1m2s
14	Gemini 3.7 Flash	google	1273	52.7%	353	±19.7	44.9s
15	Claude Opus 4.6 (Thinking)	anthropic	1269	61.9%	412	±18.0	1m11s
16	Gemini 3.5 Flash	google	1269	59.1%	1,330	±10.3	1m26s
17	Claude Opus 4.6	anthropic	1267	62.9%	377	±18.8	1m12s
18	Grok 4.5	xai	1264	53.8%	732	±13.7	56.1s
19	GPT-5.6 Sol (Medium)	openai	1263	57.2%	229	±24.2	40.9s
20	Claude Sonnet 4.6	anthropic	1249	60.1%	411	±17.9	1m26s
21	Deca 2.5 Ultra	genlabs	1240	53.9%	104	±35.0	1m5s
22	GLM-5.3	zhipu	1231	48.7%	154	±29.1	7m55s
23	Qwen3.7 Max	alibaba	1228	53.7%	751	±13.2	4m19s
24	Qwen3.8 Max	alibaba	1222	46.5%	185	±26.6	6m57s
25	Claude Sonnet 5	anthropic	1221	51.9%	638	±14.1	55.4s
26	Claude Sonnet 4.5 (Thinking)	anthropic	1220	56.1%	531	±15.6	47.5s
27	GLM 5.2	zhipu	1219	47.5%	550	±15.6	4m49s
28	GPT-5.4 (Medium)	openai	1214	55.4%	374	±18.5	1m9s
29	Kimi K2.7 Code	moonshot	1214	48.3%	460	±16.8	5m20s
30	Claude Opus 4.5	anthropic	1205	54.8%	429	±17.3	48.3s
31	GPT-5.4 (None)	openai	1204	52.0%	1,307	±10.3	29.5s
32	GPT-5.4 (Low)	openai	1203	52.4%	700	±13.7	34.4s
33	GPT-5.6 Terra	openai	1193	46.4%	1,350	±10.6	16.1s
34	Claude Sonnet 4.5	anthropic	1190	51.7%	420	±17.4	56.9s
35	Grok 4.20 Beta (Reasoning)	xai	1190	48.2%	2,039	±8.4	1m33s
36	Claude Opus 4.1	anthropic	1182	51.5%	458	±16.7	44.5s
37	Gemini 3.1 Flash-Lite Preview	google	1182	50.7%	1,209	±10.7	4.8s
38	Kimi K2.5 (Thinking)	moonshot	1181	46.3%	930	±12.0	4m46s
39	GLM 4.7	zhipu	1180	48.1%	1,134	±11.0	2m37s
40	Kimi K2.6	moonshot	1180	47.5%	120	±32.1	5m58s
41	MiniMax M3	minimax	1177	47.1%	480	±16.4	1m12s
42	GLM 5 Turbo	zhipu	1171	49.6%	345	±19.1	5m9s
43	MiMo-V2.5-Pro	xiaomi	1171	47.4%	234	±23.2	2m3s
44	GPT-5.3 Codex	openai	1167	51.2%	258	±22.2	24.0s
45	GPT-5.2 (Low)	openai	1167	50.7%	371	±18.5	45.8s
46	Step 3.7 Flash	stepfun	1165	44.9%	566	±15.1	1m48s
47	DeepSeek-V4-Pro	deepseek	1164	46.4%	679	±13.9	4m37s
48	GLM 5	zhipu	1162	47.9%	984	±11.7	1m23s
49	GPT-5.6 Luna	openai	1161	42.3%	234	±23.8	14.3s
50	MiMo-V2.5	xiaomi	1159	46.3%	320	±19.9	2m7s
51	MiniMax M2.7	minimax	1156	47.5%	512	±15.8	1m22s
52	GPT-5 (Minimal)	openai	1156	49.0%	312	±20.1	1m25s
53	Claude Haiku 4.5	anthropic	1155	49.3%	353	±19.1	11.2s
54	Grok 4.3	xai	1152	44.8%	520	±15.7	29.3s
55	Hy3	tencent	1151	40.6%	138	±31.4	4m14s
56	Qwen3.7 Plus	alibaba	1147	40.8%	196	±25.7	3m59s
57	GLM 5.1	zhipu	1144	43.9%	221	±24.0	5m25s
58	Qwen3 Max	alibaba	1143	47.2%	305	±20.4	1m17s
59	GPT-5.2 (Medium)	openai	1142	47.0%	421	±17.4	42.6s
60	Grok 4.20 Beta	xai	1141	46.3%	529	±15.6	13.1s
61	GPT-5 mini (Default)	openai	1138	44.7%	414	±17.6	39.7s
62	Qwen3.6 Plus	alibaba	1135	43.1%	390	±18.2	2m27s
63	GPT-5.1 (None)	openai	1134	48.6%	140	±29.9	26.4s
64	GPT-5.1 (Low)	openai	1131	47.0%	115	±33.0	30.4s
65	GPT-5.2 (None)	openai	1129	44.8%	395	±18.0	46.0s
66	GLM 5V Turbo	zhipu	1127	43.0%	412	±17.6	1m46s
67	DeepSeek-V4-Flash	deepseek	1126	42.8%	297	±20.8	1m52s
68	Grok 4 Fast (Reasoning)	xai	1124	45.0%	269	±21.9	18.5s
69	Grok 4.1 Fast	xai	1122	46.1%	273	±21.6	15.7s
70	GPT-5.1 Codex Mini	openai	1120	43.0%	309	±20.4	16.2s
71	GPT-5.1 (Medium)	openai	1116	46.2%	106	±34.3	27.7s
72	Nex N2 Pro	nex-agi	1114	37.4%	131	±31.9	3m29s
73	Inkling	thinkingmachines	1110	33.7%	166	±29.5	1m16s
74	Qwen3.5 Plus 02-15	alibaba	1109	43.2%	296	±20.9	2m21s
75	DeepSeek-V4-Flash-0731	deepseek	1107	33.3%	105	±36.9	5m20s
76	DeepSeek-V3.2	deepseek	1098	40.5%	259	±22.4	1m58s
77	Nemotron 3 Ultra	nvidia	1098	37.2%	218	±24.8	1m42s
78	Mistral Large 3 (2512)	mistral	1084	40.3%	340	±19.8	1m5s
79	Mistral Medium 3.5	mistral	1078	36.6%	205	±25.6	17.9s
80	GPT-5.1 (High)	openai	1076	40.2%	107	±35.0	37.6s
81	Grok 4 Fast	xai	1072	37.8%	254	±23.1	21.9s
82	Grok 4.1 Fast (Reasoning)	xai	1070	38.0%	274	±22.1	21.7s
83	Trinity Large Thinking	arceeai	1056	37.1%	116	±33.8	5m11s
84	AesCoder-4B	designflow	1054	35.1%	279	±22.3	43.0s
85	MiMo-V2-Flash	xiaomi	1032	33.9%	251	±23.7	2m45s
86	Mercury 2	inception	1017	26.5%	170	±30.6	17.8s
87	Mistral Medium 3.1 (2508)	mistral	1013	30.8%	357	±20.3	23.3s
88	Prime Intellect: INTELLECT-3	primeintellect	976	26.5%	185	±29.4	6m18s