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

3D

153 models · 281,581 votes

#	Model	Provider	ELO ▼	Win%	Battles	CI 95%	Avg Time
1	GPT-6 Astra (xhigh)	openai	1475	73.8%	1,449	±0.0	10m7s
2	Claude Fable 5.1	anthropic	1423	68.2%	1,121	±0.0	5m4s
3	Muse Spark 1.3 Max	meta	1423	64.1%	1,948	±0.0	3m38s
4	Kimi K3	moonshot	1422	68.5%	3,059	±0.0	10m25s
5	GPT-5.6 Sol (xhigh)	openai	1409	64.5%	1,475	±0.0	4m52s
6	DeepSeek-V4.1-Flash	deepseek	1393	60.4%	2,886	±0.0	2m28s
7	GLM-5.3	zhipu	1388	66.5%	1,165	±0.0	8m41s
8	Muse Spark 1.3 (xhigh)	meta	1382	59.0%	3,249	±0.0	2m24s
9	Qwen3.8 Max	alibaba	1364	58.5%	2,066	±0.0	10m21s
10	Claude Opus 5	anthropic	1359	61.1%	2,913	±0.0	5m29s
11	GLM-5.3-Flash	zhipu	1350	58.4%	4,688	±0.0	8m8s
12	MiMo-V2.6-Pro	xiaomi	1348	58.7%	1,999	±0.0	3m3s
13	Claude Fable 5	anthropic	1346	61.7%	4,418	±0.0	2m15s
14	GPT-5.6 Sol (Medium)	openai	1342	59.2%	5,296	±0.0	1m58s
15	Gemini 3.7 Flash	google	1335	55.5%	5,187	±0.0	1m35s
16	Muse Spark 1.2	meta	1333	54.0%	4,332	±0.0	3m35s
17	Gemini 3.8 Flash	google	1324	51.3%	3,801	±0.0	1m31s
18	GLM 5.2	zhipu	1319	53.7%	7,436	±0.0	6m4s
19	Deca 2.6 Ultra	genlabs	1310	54.0%	437	±0.0	3m59s
20	Grok 4.6	xai	1304	52.7%	8,459	±0.0	3m53s
21	Claude Opus 4.6	anthropic	1302	62.5%	5,859	±0.0	2m36s
22	Gemini 3.6 Flash	google	1300	53.1%	7,312	±0.0	1m5s
23	Claude Opus 4.6 (Thinking)	anthropic	1299	60.5%	5,086	±0.0	2m42s
24	Kimi K2.6	moonshot	1296	57.6%	13,976	±0.0	5m40s
25	Qwen3.7 Max	alibaba	1295	55.1%	6,394	±0.0	3m5s
26	Muse Spark 1.1	meta	1288	50.1%	4,659	±0.0	3m39s
27	Claude Sonnet 5	anthropic	1287	54.5%	4,830	±0.0	3m35s
28	Grok 4.5	xai	1287	50.5%	11,020	±0.0	2m3s
29	DeepSeek-V4-Pro	deepseek	1282	56.8%	4,509	±0.0	5m4s
30	MiMo-V2.5-Pro	xiaomi	1278	55.0%	3,798	±0.0	3m29s
31	Nex N2 Pro	nex-agi	1277	52.9%	3,857	±0.0	6m47s
32	GLM 5 Turbo	zhipu	1276	56.4%	5,244	±0.0	5m32s
33	GLM 5.1	zhipu	1275	53.4%	5,792	±0.0	6m30s
34	Gemini 3.5 Flash	google	1270	56.4%	6,814	±0.0	1m11s
35	Kimi K2.7 Code	moonshot	1270	51.0%	6,092	±0.0	4m12s
36	Deca 2.5 Ultra	genlabs	1268	53.0%	168	±0.0	1m22s
37	Qwen3.7 Plus	alibaba	1267	48.6%	4,436	±0.0	2m56s
38	Claude Sonnet 4.6	anthropic	1262	57.6%	5,715	±0.0	2m13s
39	Deca 2.5 Ultra	genlabs	1262	50.7%	835	±0.0	4m57s
40	Gemini 3.1 Pro Preview	google	1262	58.4%	9,104	±0.0	1m29s
41	Hy3	tencent	1258	50.4%	3,580	±0.0	3m57s
42	GLM 5	zhipu	1256	56.3%	9,255	±0.0	2m15s
43	Claude Opus 4.8	anthropic	1251	52.5%	6,746	±0.0	1m25s
44	Gemini 3 Pro Preview	google	1251	66.7%	2,519	±0.0	1m13s
45	Muse Spark	meta	1248	63.5%	561	±0.0	2m8s
46	MiMo-V2.5	xiaomi	1242	51.0%	10,475	±0.0	2m7s
47	Claude Opus 4.5	anthropic	1241	58.5%	6,015	±0.0	1m26s
48	GLM 5V Turbo	zhipu	1237	53.0%	12,778	±0.0	1m50s
49	DeepSeek-V4-Flash-0731	deepseek	1236	48.6%	1,226	±0.0	4m15s
50	MiniMax M3	minimax	1235	50.4%	4,575	±0.0	1m40s
51	Kimi K2.5 (Thinking)	moonshot	1233	53.1%	11,631	±0.0	2m28s
52	Qwen3.6 Plus	alibaba	1230	50.9%	12,170	±0.0	2m52s
53	GPT-5.5	openai	1225	50.3%	5,125	±0.0	2m19s
54	MiniMax M2.7	minimax	1221	50.0%	8,230	±0.0	2m8s
55	Grok 4.20 Beta (Reasoning)	xai	1221	52.1%	9,940	±0.0	1m40s
56	Inkling	thinkingmachines	1219	47.0%	232	±0.0	4m10s
57	GLM 4.7	zhipu	1218	54.3%	7,003	±0.0	2m23s
58	DeepSeek-V4-Flash	deepseek	1216	49.3%	8,026	±0.0	1m47s
59	Gemini 3 Flash Preview	google	1211	62.7%	467	±0.0	24.4s
60	Solar Pro 4	upstage	1207	46.4%	2,686	±0.0	4m19s
61	Hy3 Low	tencent	1203	43.9%	764	±0.0	2m1s
62	GLM 4.5	zhipu	1198	59.7%	1,878	±0.0	1m49s
63	MiniMax M2.5	minimax	1197	57.6%	1,289	±0.0	1m33s
64	MiniMax M2.1	minimax	1193	57.5%	2,159	±0.0	1m36s
65	Qwen3.5 397B A17B	alibaba	1188	56.7%	960	±0.0	1m45s
66	Inkling	thinkingmachines	1186	40.4%	1,657	±0.0	57.6s
67	Claude Sonnet 4.5	anthropic	1185	51.0%	7,374	±0.0	1m6s
68	Claude Opus 4.1	anthropic	1181	51.9%	5,944	±0.0	1m17s
69	Claude Sonnet 4.5 (Thinking)	anthropic	1180	50.3%	7,304	±0.0	1m3s
70	Grok 4.20 Beta	xai	1179	46.9%	10,935	±0.0	51.7s
71	DeepSeek-V3.2-Exp	deepseek	1176	56.4%	1,976	±0.0	2m32s
72	AGI-01 Swift	lucidquery	1168	43.8%	480	±0.0	1m3s
73	Claude Opus 4.1 (Thinking)	anthropic	1168	56.5%	1,291	±0.0	1m17s
74	DeepSeek-V3.1 (Thinking)	deepseek	1168	56.0%	1,456	±0.0	2m36s
75	Claude Sonnet 4	anthropic	1167	57.8%	1,574	±0.0	57.0s
76	Claude Opus 4	anthropic	1166	57.7%	1,469	±0.0	1m7s
77	GPT-5.6 Terra	openai	1163	37.2%	16,752	±0.0	47.1s
78	DeepSeek-V3.2	deepseek	1160	49.4%	4,159	±0.0	2m13s
79	Nemotron 3 Ultra	nvidia	1160	39.7%	3,662	±0.0	1m49s
80	GLM 4.6	zhipu	1156	54.2%	1,671	±0.0	1m20s
81	Step 3.7 Flash	stepfun	1155	40.9%	6,387	±0.0	1m45s
82	Claude Sonnet 4 (Thinking)	anthropic	1154	55.0%	1,463	±0.0	53.6s
83	Grok 4.3	xai	1153	41.6%	10,990	±0.0	46.5s
84	GPT-5.6 Luna	openai	1151	38.0%	3,153	±0.0	32.8s
85	GLM 4.5 Air	zhipu	1150	54.1%	1,475	±0.0	1m27s
86	GLM 4.7 Flash	zhipu	1148	51.2%	1,460	±0.0	1m57s
87	Qwen3.5 Plus 02-15	alibaba	1147	47.7%	2,648	±0.0	1m37s
88	MiMo-V2-Flash	xiaomi	1143	49.2%	4,314	±0.0	44.8s
89	DeepSeek-R1-0528	deepseek	1139	53.3%	1,630	±0.0	1m9s
90	GPT-5.4 (Medium)	openai	1129	42.3%	2,925	±0.0	3m16s
91	Mistral Large 3 (2512)	mistral	1127	46.9%	4,549	±0.0	1m7s
92	MiniMax M2 Stable	minimax	1125	48.3%	1,322	±0.0	1m41s
93	Claude 3.7 Sonnet	anthropic	1122	51.9%	1,371	±0.0	1m19s
94	v0-1.5-md	vercel	1118	50.2%	1,089	±0.0	42.7s
95	DeepSeek-V3-0324	deepseek	1114	50.5%	1,821	±0.0	43.0s
96	Prime Intellect: INTELLECT-3	primeintellect	1114	45.2%	3,897	±0.0	59.1s
97	Mistral Medium 3 (2505)	mistral	1111	54.7%	512	±0.0	51.4s
98	AesCoder-4B	designflow	1109	45.1%	4,814	±0.0	36.6s
99	Mistral Medium 3.1 (2508)	mistral	1109	44.6%	4,180	±0.0	1m19s
100	Trinity Large Thinking	arceeai	1109	41.3%	2,117	±0.0	1m8s
101	Gemini 2.5 Pro	google	1108	50.6%	644	±0.0	57.1s
102	GPT-5.2 (Medium)	openai	1107	41.5%	4,921	±0.0	2m18s
103	DeepSeek-V3.1	deepseek	1104	47.9%	1,757	±0.0	28.9s
104	Qwen3 Max	alibaba	1101	43.5%	4,044	±0.0	1m13s
105	Claude Haiku 4.5	anthropic	1100	41.1%	6,343	±0.0	29.7s
106	GPT-5.2 (None)	openai	1100	40.8%	5,007	±0.0	2m11s
107	Gemini 2.5 Flash Preview 09-2025	google	1099	47.4%	1,587	±0.0	22.4s
108	GPT-5.2 (Low)	openai	1094	42.0%	3,742	±0.0	2m7s
109	GPT-5.4 (Low)	openai	1093	38.9%	4,296	±0.0	1m5s
110	Grok Code Fast 1	xai	1092	49.5%	291	±0.0	13.8s
111	Grok 4.1 Fast	xai	1089	42.9%	4,167	±0.0	24.1s
112	GPT-5.1 (High)	openai	1088	43.7%	1,850	±0.0	2m30s
113	Mistral Medium 3.5	mistral	1088	38.3%	2,065	±0.0	31.4s
114	GPT-5 (Minimal)	openai	1083	41.3%	3,973	±0.0	1m57s
115	Grok 4 Fast	xai	1083	41.7%	4,571	±0.0	31.3s
116	GPT-5.2 (High)	openai	1081	45.3%	342	±0.0	2m0s
117	GPT-5.1 (None)	openai	1078	43.0%	2,361	±0.0	1m22s
118	Grok 4	xai	1078	44.3%	2,454	±0.0	52.6s
119	Grok 4 Fast (Reasoning)	xai	1078	41.2%	4,739	±0.0	21.0s
120	Grok 3	xai	1077	44.3%	2,656	±0.0	27.2s
121	Gemini 3.1 Flash-Lite Preview	google	1075	38.7%	3,832	±0.0	4.5s
122	GPT-5.4 (None)	openai	1075	37.0%	4,648	±0.0	55.0s
123	GPT-5.1 (Low)	openai	1070	42.0%	2,355	±0.0	1m14s
124	GPT-5 mini (Default)	openai	1065	36.9%	5,577	±0.0	1m41s
125	GPT-5.1 (Medium)	openai	1065	41.4%	2,300	±0.0	1m35s
126	Magistral Medium 1.2 (2509)	mistral	1061	45.0%	389	±0.0	1m17s
127	Devstral Medium	mistral	1060	48.1%	478	±0.0	17.0s
128	Ministral 3 8B (2512)	mistral	1059	46.2%	238	±0.0	33.2s
129	GPT-5 (High)	openai	1054	39.6%	1,149	±0.0	2m34s
130	Grok 4.1 Fast (Reasoning)	xai	1049	37.3%	3,889	±0.0	57.6s
131	Gemini 2.5 Flash	google	1048	44.4%	495	±0.0	45.2s
132	Codestral 2508	mistral	1047	45.4%	445	±0.0	7.8s
133	GPT-5.3 Codex	openai	1036	35.3%	2,219	±0.0	46.0s
134	Magistral Small 1.2 (2509)	mistral	1034	41.3%	499	±0.0	22.8s
135	Olmo 3.1 32B Think	allenai	1030	35.6%	1,724	±0.0	1m49s
136	Qwen3-235B-A22B-Thinking-2507	alibaba	1025	40.4%	384	±0.0	1m28s
137	Qwen3-235B-A22B-Instruct-2507	alibaba	1022	41.1%	452	±0.0	39.5s
138	Ministral 3 14B (2512)	mistral	1017	39.6%	212	±0.0	53.6s
139	GPT-5.1 Codex Mini	openai	1016	32.8%	4,527	±0.0	34.4s
140	Llama 3.1 Nemotron Ultra 253B	nvidia	1010	39.0%	228	±0.0	21.5s
141	Mercury 2	inception	999	21.6%	3,271	±0.0	4.2s
142	Gemini 2.5 Flash Lite Preview 09-2025	google	993	36.3%	413	±0.0	6.5s
143	GPT-5 nano (Default)	openai	993	36.3%	383	±0.0	1m28s
144	Ministral 3 3B (2512)	mistral	990	35.9%	195	±0.0	29.2s
145	Grok 3 Mini	xai	983	41.5%	557	±0.0	18.4s
146	GPT-4.1 nano	openai	952	46.0%	213	±0.0	17.1s
147	GPT OSS 120B	openai	928	29.4%	313	±0.0	16.9s
148	Llama 4 Maverick	meta	928	40.2%	164	±0.0	9.5s
149	GPT-4o	openai	896	39.2%	194	±0.0	11.5s
150	Qwen3-235B-A22B	alibaba	886	24.5%	421	±0.0	1m8s
151	o4-mini	openai	881	34.0%	188	±0.0	30.6s
152	GPT-4.1	openai	877	30.9%	181	±0.0	1m8s
153	GPT-4.1 mini	openai	864	30.5%	151	±0.0	1m2s