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

ASCII

70 models · 14,619 votes

#	Model	Provider	ELO ▼	Win%	Battles	CI 95%	Avg Time
1	Claude Fable 5	anthropic	1364	69.9%	166	±0.0	1m27s
2	Claude Opus 4.7	anthropic	1323	66.7%	345	±0.0	40.1s
3	GPT-5.6 Sol	openai	1312	60.6%	104	±0.0	39.3s
4	Gemini 3.1 Pro Preview	google	1308	63.9%	1,253	±0.0	3m46s
5	Grok 4.5	xai	1301	59.0%	105	±0.0	50.0s
6	Claude Opus 4.8	anthropic	1300	62.5%	275	±0.0	26.2s
7	Gemini 3.5 Flash	google	1299	61.1%	997	±0.0	1m21s
8	Claude Opus 4.6	anthropic	1291	63.6%	360	±0.0	1m13s
9	Claude Opus 4.6 (Thinking)	anthropic	1289	61.8%	393	±0.0	1m11s
10	GPT-5.5	openai	1286	60.8%	332	±0.0	1m6s
11	Claude Sonnet 4.6	anthropic	1267	59.6%	396	±0.0	1m25s
12	Qwen3.7 Max	alibaba	1254	55.0%	582	±0.0	4m6s
13	Claude Sonnet 4.5 (Thinking)	anthropic	1242	56.3%	506	±0.0	46.0s
14	GPT-5.4 (Medium)	openai	1237	55.5%	373	±0.0	1m8s
15	GLM 5.2	zhipu	1237	51.7%	145	±0.0	4m50s
16	GPT-5.4 (None)	openai	1228	51.9%	1,304	±0.0	29.4s
17	Claude Opus 4.5	anthropic	1227	54.7%	426	±0.0	48.3s
18	GPT-5.4 (Low)	openai	1226	52.4%	700	±0.0	34.4s
19	Kimi K2.7 Code	moonshot	1218	50.6%	158	±0.0	5m35s
20	Grok 4.20 Beta (Reasoning)	xai	1215	49.4%	1,689	±0.0	1m35s
21	Claude Sonnet 4.5	anthropic	1213	52.0%	398	±0.0	56.5s
22	Kimi K2.5 (Thinking)	moonshot	1206	47.1%	772	±0.0	4m44s
23	Gemini 3.1 Flash-Lite Preview	google	1205	50.7%	1,203	±0.0	4.8s
24	GLM 4.7	zhipu	1204	48.2%	1,131	±0.0	2m37s
25	Claude Opus 4.1	anthropic	1203	51.4%	457	±0.0	44.5s
26	MiniMax M3	minimax	1194	47.3%	372	±0.0	1m16s
27	Step 3.7 Flash	stepfun	1193	46.8%	489	±0.0	1m42s
28	GLM 5 Turbo	zhipu	1191	50.0%	294	±0.0	5m1s
29	GPT-5.2 (Low)	openai	1190	50.7%	371	±0.0	45.8s
30	GPT-5.3 Codex	openai	1189	51.2%	258	±0.0	24.0s
31	DeepSeek-V4-Pro	deepseek	1188	46.6%	637	±0.0	4m46s
32	Kimi K2.6	moonshot	1188	46.0%	100	±0.0	6m7s
33	GLM 5	zhipu	1186	48.0%	982	±0.0	1m23s
34	Qwen3.7 Plus	alibaba	1185	45.5%	123	±0.0	4m16s
35	GLM 5.1	zhipu	1184	47.8%	157	±0.0	5m15s
36	Grok 4.3	xai	1180	46.4%	433	±0.0	32.5s
37	MiMo-V2.5-Pro	xiaomi	1180	47.1%	172	±0.0	2m2s
38	GPT-5 (Minimal)	openai	1179	49.0%	312	±0.0	1m25s
39	Claude Haiku 4.5	anthropic	1178	49.3%	353	±0.0	11.2s
40	MiniMax M2.7	minimax	1178	48.1%	451	±0.0	1m24s
41	MiMo-V2.5	xiaomi	1170	45.7%	245	±0.0	2m14s
42	Grok 4.20 Beta	xai	1168	48.3%	441	±0.0	13.4s
43	Qwen3 Max	alibaba	1166	47.2%	305	±0.0	1m17s
44	GPT-5.2 (Medium)	openai	1164	47.0%	421	±0.0	42.6s
45	Qwen3.6 Plus	alibaba	1161	45.4%	315	±0.0	2m28s
46	GPT-5 mini (Default)	openai	1159	44.5%	411	±0.0	39.7s
47	GPT-5.1 (None)	openai	1156	48.6%	140	±0.0	26.4s
48	GPT-5.1 (Low)	openai	1154	47.0%	115	±0.0	30.4s
49	DeepSeek-V4-Flash	deepseek	1153	43.3%	289	±0.0	1m50s
50	GPT-5.2 (None)	openai	1150	44.8%	395	±0.0	46.0s
51	Grok 4.1 Fast	xai	1147	46.5%	271	±0.0	15.8s
52	Grok 4 Fast (Reasoning)	xai	1146	45.0%	269	±0.0	18.5s
53	GPT-5.1 Codex Mini	openai	1142	43.0%	309	±0.0	16.2s
54	GPT-5.1 (Medium)	openai	1138	46.2%	106	±0.0	27.7s
55	GLM 5V Turbo	zhipu	1137	42.6%	324	±0.0	2m2s
56	Qwen3.5 Plus 02-15	alibaba	1132	43.2%	296	±0.0	2m21s
57	Nex N2 Pro	nex-agi	1131	37.3%	102	±0.0	4m4s
58	DeepSeek-V3.2	deepseek	1120	40.5%	259	±0.0	1m58s
59	Nemotron 3 Ultra	nvidia	1111	37.5%	144	±0.0	1m35s
60	Mistral Large 3 (2512)	mistral	1107	40.3%	340	±0.0	1m5s
61	Mistral Medium 3.5	mistral	1100	36.6%	205	±0.0	17.9s
62	GPT-5.1 (High)	openai	1099	40.2%	107	±0.0	37.6s
63	Grok 4 Fast	xai	1095	37.9%	253	±0.0	22.0s
64	Grok 4.1 Fast (Reasoning)	xai	1093	38.0%	274	±0.0	21.7s
65	Trinity Large Thinking	arceeai	1079	37.1%	116	±0.0	5m11s
66	AesCoder-4B	designflow	1077	35.1%	279	±0.0	43.0s
67	MiMo-V2-Flash	xiaomi	1055	33.9%	251	±0.0	2m45s
68	Mercury 2	inception	1043	28.2%	103	±0.0	10.7s
69	Mistral Medium 3.1 (2508)	mistral	1036	30.8%	357	±0.0	23.3s
70	Prime Intellect: INTELLECT-3	primeintellect	998	26.5%	185	±0.0	6m18s