Arcada Labs Code Categories Arena 代码能力排行榜

Name: Arcada Labs Code Categories Arena 代码能力排行榜
Creator: DataLearner
License: https://creativecommons.org/licenses/by/4.0/

基于 Arcada Labs Code Categories Arena 用户匿名投票的最新AI大模型代码能力排行榜，通过 Bradley-Terry 模型对 Website、UI Component、Game Dev、Data Visualization 等代码子类别进行综合评分与排名。

榜首模型

GLM 5.2

最高得分

1352.00

模型数量

136

数据版本

2026年07月12日

数据来源: Arcada Labs

来源：全部国产模型

榜单历史快照月份:

排名总表

排名	模型名称	得分	95% CI	投票数	机构	许可证
	GLM 5.2Zhipu AI	1352.00	+/-7.9	8,533	Zhipu AI	Open Source
	GPT-5.6 SolOpenAI	1350.00	+/-16.8	1,792	OpenAI	Proprietary
	Claude Fable 5Anthropic	1344.00	+/-11.6	3,881	Anthropic	Proprietary
4	Claude Opus 4.6Anthropic	1336.00	+/-5.1	23,055	Anthropic	Proprietary
5	Claude Opus 4.6 (thinking)Anthropic	1330.00	+/-5.5	19,041	Anthropic	Proprietary
6	Opus 4.7Anthropic	1330.00	+/-5.8	16,847	Anthropic	Proprietary
7	Grok 4.5xAI	1328.00	+/-11.9	3,493	xAI	Proprietary
8	GLM 5.1智谱AI	1321.00	+/-6.6	12,099	智谱AI	Open Source
9	Kimi K2.6Moonshot AI	1320.00	+/-4.9	23,919	Moonshot AI	Open Source
10	Claude Sonnet 4.6Anthropic	1319.00	+/-5.1	22,045	Anthropic	Proprietary
11	GLM-5-Turbo智谱AI	1312.00	+/-4.7	27,317	智谱AI	Proprietary
12	MiMo-V2.5-ProXiaomi	1312.00	+/-7.2	10,020	Xiaomi	Open Source
13	Qwen3.7 Max阿里巴巴	1306.00	+/-5.8	16,473	阿里巴巴	Proprietary
14	Kimi K2.7 CodeMoonshot AI	1300.00	+/-8.2	7,614	Moonshot AI	Open Source
15	Gemini 3.5 FlashGoogle Deep Mind	1297.00	+/-5.8	16,254	Google Deep Mind	Proprietary
16	MiniMax M3MiniMax	1297.00	+/-6.5	12,553	MiniMax	Open Source
17	MiMo-V2.5Xiaomi	1295.00	+/-4.4	31,718	Xiaomi	Open Source
18	Muse SparkFacebook AI研究实验室	1294.00	+/-10.9	4,249	Facebook AI研究实验室	Proprietary
19	GPT-5.5OpenAI	1291.00	+/-5.9	15,610	OpenAI	Proprietary
20	GLM-5智谱AI	1288.00	+/-3.8	46,436	智谱AI	Open Source
21	Opus 4.5Anthropic	1284.00	+/-4.1	35,670	Anthropic	Proprietary
22	Gemini 3.1 Pro PreviewGoogle Deep Mind	1283.00	+/-5	23,843	Google Deep Mind	Proprietary
23	DeepSeek-V4-ProDeepSeek-AI	1281.00	+/-5.5	17,824	DeepSeek-AI	Open Source
24	Kimi K2.5 (thinking)Moonshot AI	1279.00	+/-4	39,455	Moonshot AI	Open Source
25	Claude Opus 4.8Anthropic	1276.00	+/-5.9	15,114	Anthropic	Proprietary
26	MiniMax-M2.7MiniMaxAI	1275.00	+/-4.4	31,074	MiniMaxAI	Open Source
27	Gemini 3.1 Pro PreviewGoogle Deep Mind	1271.00	+/-4.2	35,060	Google Deep Mind	Proprietary
28	GLM-5V-Turbo智谱AI	1270.00	+/-4.3	31,672	智谱AI	Proprietary
29	Nex N2 ProNex AGI	1265.00	+/-9.1	5,932	Nex AGI	Open Source
30	Qwen 3.6 Plus Preview阿里巴巴	1265.00	+/-4.7	26,080	阿里巴巴	Proprietary
31	Grok 4.20 Beta (Reasoning)xAI	1260.00	+/-4.7	26,100	xAI	Proprietary
32	GLM-4.7智谱AI	1258.00	+/-3.7	47,633	智谱AI	Open Source
33	GPT-5.4 (Design Skill, Medium)OpenAI	1254.00	+/-7.2	9,880	OpenAI	Proprietary
34	GPT-5.4 (medium)OpenAI	1252.00	+/-5.3	19,407	OpenAI	Proprietary
35	MiniMax M2.5MiniMaxAI	1249.00	+/-6.7	11,504	MiniMaxAI	Open Source
36	DeepSeek-V4-FlashDeepSeek-AI	1248.00	+/-4.7	25,437	DeepSeek-AI	Open Source
37	Grok 4.20 BetaxAI	1241.00	+/-4.6	27,379	xAI	Proprietary
38	Grok 4.3 BetaxAI	1234.00	+/-5.1	21,060	xAI	Proprietary
39	M2.1MiniMaxAI	1232.00	+/-5.1	20,805	MiniMaxAI	Open Source
40	Gemini 3.0 FlashGoogle Deep Mind	1231.00	+/-10.6	4,414	Google Deep Mind	Proprietary
41	Claude Sonnet 4.5Anthropic	1225.00	+/-3.8	40,900	Anthropic	Proprietary
42	Claude Sonnet 4.5 (thinking)Anthropic	1224.00	+/-3.9	40,107	Anthropic	Proprietary
43	GPT-5.4 (low)OpenAI	1222.00	+/-5	21,683	OpenAI	Proprietary
44	Qwen3.5-397B-A17B阿里巴巴	1222.00	+/-7.9	8,131	阿里巴巴	Open Source
45	GLM-4.7-Flash智谱AI	1220.00	+/-6.6	11,706	智谱AI	Open Source
46	GPT-5.4 (None)OpenAI	1220.00	+/-4.8	24,150	OpenAI	Proprietary
47	Claude Sonnet 3.7Anthropic	1219.00	+/-5.9	15,245	Anthropic	Proprietary
48	DeepSeek-V3.1 (thinking)DeepSeek-AI	1217.00	+/-5.7	16,258	DeepSeek-AI	Open Source
49	Opus 4.1 (thinking)Anthropic	1213.00	+/-5.8	15,677	Anthropic	Proprietary
50	DeepSeek V3.2-ExpDeepSeek-AI	1213.00	+/-5.3	19,490	DeepSeek-AI	Open Source
51	GPT-5.1 (high)OpenAI	1213.00	+/-5.7	16,057	OpenAI	Proprietary
52	GPT-5.2 (medium)OpenAI	1212.00	+/-4.5	28,372	OpenAI	Proprietary
53	GPT-5 (high)OpenAI	1211.00	+/-6.2	13,397	OpenAI	Proprietary
54	GPT-5.2 (None)OpenAI	1211.00	+/-4.4	29,322	OpenAI	Proprietary
55	Step 3.7 FlashStepFun	1210.00	+/-6.3	13,365	StepFun	Open Source
56	Qwen3.5 Plus (0215)阿里巴巴	1209.00	+/-5.3	19,028	阿里巴巴	Proprietary
57	GLM-4.6智谱AI	1208.00	+/-5.6	16,911	智谱AI	Open Source
58	GPT-5.2 (low)OpenAI	1208.00	+/-4.6	25,763	OpenAI	Proprietary
59	Opus 4.1Anthropic	1207.00	+/-3.9	40,219	Anthropic	Proprietary
60	GLM-4.5智谱AI	1207.00	+/-5.2	19,637	智谱AI	Open Source
61	GPT-5 (minimal)OpenAI	1207.00	+/-4.2	33,251	OpenAI	Proprietary
62	DeepSeek V3.2DeepSeek-AI	1206.00	+/-4.4	29,394	DeepSeek-AI	Open Source
63	GPT-5.1 (medium)OpenAI	1204.00	+/-5	21,292	OpenAI	Proprietary
64	Claude Opus 4Anthropic	1203.00	+/-5.6	16,669	Anthropic	Proprietary
65	Hy3Tencent	1199.00	+/-20.5	1,202	Tencent	Open Source
66	GPT-5.1 (low)OpenAI	1198.00	+/-5	22,160	OpenAI	Proprietary
67	MiMo-V2-FlashXiaomi	1198.00	+/-4.1	35,857	Xiaomi	Open Source
68	Gemini 2.5-ProGoogle Deep Mind	1196.00	+/-8.6	7,044	Google Deep Mind	Proprietary
69	GPT-5.1 CodexOpenAI	1193.00	+/-16.4	1,807	OpenAI	Proprietary
70	GPT-5.1 (None)OpenAI	1193.00	+/-4.9	22,273	OpenAI	Proprietary
71	GPT-5.2 (high)OpenAI	1192.00	+/-10.8	4,167	OpenAI	Proprietary
72	GPT-5.3 CodexOpenAI	1187.00	+/-5.8	15,761	OpenAI	Proprietary
73	Qwen3-Coder-480B-A35B阿里巴巴	1185.00	+/-16.3	1,958	阿里巴巴	Open Source
74	Mistral Large 3MistralAI	1184.00	+/-4.3	30,837	MistralAI	Open Source
75	Claude Sonnet 4Anthropic	1183.00	+/-5.5	17,532	Anthropic	Proprietary
76	DeepSeek-R1-0528DeepSeek-AI	1181.00	+/-5.4	17,944	DeepSeek-AI	Open Source
77	GLM-4.5-Air智谱AI	1180.00	+/-5.5	17,256	智谱AI	Open Source
78	Claude Sonnet 4 (thinking)Anthropic	1179.00	+/-5.7	16,227	Anthropic	Proprietary
79	MiniMax M2MiniMaxAI	1177.00	+/-6.9	10,828	MiniMaxAI	Open Source
80	AesCoder-4BDesignFlow	1167.00	+/-3.9	40,177	DesignFlow	Open Source
81	Mistral Medium 3.5MistralAI	1164.00	+/-6.8	11,490	MistralAI	Open Source
82	Nemotron 3 UltraNVIDIA	1164.00	+/-7.6	9,540	NVIDIA	Open Source
83	Mistral Medium 3.1 (2508)Mistral	1163.00	+/-4.5	28,014	Mistral	Proprietary
84	Trinity Large ThinkingArcee AI	1158.00	+/-6.4	13,416	Arcee AI	Open Source
85	Haiku 4.5Anthropic	1156.00	+/-4.1	36,008	Anthropic	Proprietary
86	GPT-5-miniOpenAI	1156.00	+/-4	37,565	OpenAI	Proprietary
87	DeepSeek-V3.1DeepSeek-AI	1154.00	+/-5.1	20,278	DeepSeek-AI	Open Source
88	Qwen3-Max-Thinking阿里巴巴	1152.00	+/-4.2	33,808	阿里巴巴	Proprietary
89	DeepSeek-V3-0324DeepSeek-AI	1151.00	+/-5.3	19,257	DeepSeek-AI	Open Source
90	Prime Intellect: INTELLECT-3Prime Intellect	1149.00	+/-4.3	31,884	Prime Intellect	Open Source
91	Gemini 2.5 Flash-Preview-09-2025Google Deep Mind	1146.00	+/-5.3	19,299	Google Deep Mind	Proprietary
92	Grok 4 FastxAI	1143.00	+/-4	37,255	xAI	Proprietary
93	Kimi K2 0905Moonshot AI	1140.00	+/-17.9	1,504	Moonshot AI	Open Source
94	GPT-5.1 Codex MiniOpenAI	1136.00	+/-4.2	34,250	OpenAI	Proprietary
95	Grok 4.1 FastxAI	1135.00	+/-4.2	33,898	xAI	Proprietary
96	Grok 4.1 Fast (reasoning)xAI	1130.00	+/-4.3	31,599	xAI	Proprietary
97	GPT-5-NanoOpenAI	1127.00	+/-8.6	6,710	OpenAI	Proprietary
98	Kimi K2 Turbo PreviewMoonshot AI	1126.00	+/-15.2	2,094	Moonshot AI	Open Source
99	Gemini 2.5 Flash-Lite-Preview-09-2025Google Deep Mind	1123.00	+/-8.5	6,860	Google Deep Mind	Proprietary
100	Gemini 3.1 Flash-Lite PreviewGoogle	1114.00	+/-5	24,146	Google	Proprietary
101	Phi-3-medium 14B-previewMicrosoft Azure	1111.00	+/-8.9	6,396	Microsoft Azure	Proprietary
102	Ministral 3 14BMistralAI	1107.00	+/-14.4	2,379	MistralAI	Open Source
103	Gemini 2.5 FlashGoogle Deep Mind	1101.00	+/-8.5	6,960	Google Deep Mind	Proprietary
104	v0-1.5-mdVercel	1099.00	+/-6.9	11,081	Vercel	Proprietary
105	Grok 3xAI	1095.00	+/-4.6	26,862	xAI	Proprietary
106	Ministral 3 8BMistralAI	1095.00	+/-14.3	2,427	MistralAI	Open Source
107	Grok 4 Fast (reasoning)xAI	1093.00	+/-4.1	37,909	xAI	Proprietary
108	Qwen3-235B-A22B-2507阿里巴巴	1081.00	+/-8.6	6,932	阿里巴巴	Open Source
109	Kimi K2Moonshot AI	1076.00	+/-19.4	1,352	Moonshot AI	Open Source
110	Magistral Medium 1.2 (2509)Mistral	1076.00	+/-9.4	5,851	Mistral	Proprietary
111	Qwen3-235B-A22B-Thinking-2507Alibaba	1075.00	+/-9.1	6,169	Alibaba	Open Source
112	GPT-4.1OpenAI	1068.00	+/-17.3	1,747	OpenAI	Proprietary
113	OpenAI o3OpenAI	1062.00	+/-19.5	1,365	OpenAI	Proprietary
114	Grok 4xAI	1058.00	+/-5	23,998	xAI	Proprietary
115	Devstral MediumMistralAI	1055.00	+/-8.6	7,158	MistralAI	Proprietary
116	Ministral 3 3B (2512)Mistral	1052.00	+/-13.5	2,852	Mistral	Open Source
117	Codestral 2508Mistral	1049.00	+/-8.8	6,745	Mistral	Proprietary
118	Qwen3-235B-A22B阿里巴巴	1044.00	+/-10.1	5,154	阿里巴巴	Open Source
119	Grok Code Fast 1xAI	1041.00	+/-11.1	4,295	xAI	Proprietary
120	GPT-4.1 miniOpenAI	1036.00	+/-18.3	1,566	OpenAI	Proprietary
121	Mercury 2Inception	1033.00	+/-10.3	6,814	Inception	Proprietary
122	Magistral Small 1.2 (2509)Mistral	1028.00	+/-9.2	6,448	Mistral	Open Source
123	OpenAI o4 - miniOpenAI	1018.00	+/-16.2	2,011	OpenAI	Proprietary
124	Olmo 3.1 32B ThinkAllen AI	1017.00	+/-6.3	16,162	Allen AI	Open Source
125	GPT-4.1 nanoOpenAI	1005.00	+/-16.9	1,901	OpenAI	Proprietary
126	GPT OSS 120BOpenAI	1005.00	+/-10.3	5,268	OpenAI	Open Source
127	Qwen3-30B-A3B阿里巴巴	984.00	+/-14.6	2,575	阿里巴巴	Open Source
128	Grok 3 minixAI	973.00	+/-8.8	7,626	xAI	Proprietary
129	Llama 3.1 Nemotron Ultra 253BNVIDIA	971.00	+/-13.8	3,172	NVIDIA	Open Source
130	Mistral-Small-3.2MistralAI	949.00	+/-20.8	1,243	MistralAI	Open Source
131	Llama 4 MaverickFacebook AI研究实验室	922.00	+/-18.4	1,678	Facebook AI研究实验室	Open Source
132	Mistral Large 2.1 (2411)Mistral	905.00	+/-21	1,317	Mistral	Proprietary
133	GPT-4oOpenAI	903.00	+/-18.1	1,780	OpenAI	Proprietary
134	Codestral 2 (2501)Mistral	876.00	+/-20.7	1,444	Mistral	Open Source
135	Devstral Small 1.1MistralAI	849.00	+/-22.5	1,250	MistralAI	Open Source
136	Llama 4 ScoutFacebook AI研究实验室	832.00	+/-22.6	1,275	Facebook AI研究实验室	Open Source

数据仅供参考，以官方来源为准。模型名称旁的链接可跳转到 DataLearner 模型详情页。

关于本榜单

本榜单数据来源于Design Arena，由 Y Combinator 支持的 Arcada Labs 开发，是专注于评测 AI 设计代码生成能力的众包匿名对战平台。

与 LMArena 评测通用文本和编程能力不同，Design Arena 的代码榜专门考察模型生成具有视觉呈现效果的前端代码的能力。平台将代码任务细分为 Website、UI 组件、游戏开发、数据可视化、SVG、Web App、移动端等多个子类别，每个子类别均有独立排行。

本页展示的是 Code Categories 综合榜，即将所有子类别的用户投票混合汇总后，统一用 Bradley-Terry 模型（类 Elo 算法）计算出的综合排名。每票等权，不对各子类别做加权处理，因此投票量较大的子类别（如 Website）对综合分数的影响更大。得分越高，代表模型在设计代码生成场景下的综合人类偏好越强。

常见问题 (FAQ)

什么是 Arcada Labs Code Categories Arena？

Arcada Labs Code Categories Arena 是专注于设计代码生成能力的匿名评测平台，覆盖 Website、UI 组件、游戏开发、数据可视化等多个代码生成子类别，并将投票汇总为综合榜单。

Arcada Code Arena 与 LMArena Coding Arena 有什么区别？

LMArena Coding Arena 主要评测通用编程能力，例如代码生成、调试和算法实现；Arcada Code Arena 专注于具有视觉呈现效果的前端设计代码，例如 HTML 页面、交互 UI、图表、SVG 和原型。

排名方法论是什么？

Arcada Labs 将各代码子类别的原始投票混合后运行 Bradley-Terry 模型。每票等权，不按子类别单独加权，因此投票量较大的子类别会对综合分数产生更大影响。

哪类模型在设计代码场景表现更好？

具备强视觉理解和前端代码生成能力的大模型通常表现更好。针对 UI 和代码生成优化的专项模型，在布局、交互和视觉细节任务上也可能有突出表现。