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

Kimi K3

最高得分

1419.00

模型数量

145

数据版本

2026年07月26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
	Kimi K3Moonshot AI	1419.00	+/-11.7	4,034	Moonshot AI	Open Source
	GPT-5.6 SolOpenAI	1359.00	+/-10.1	5,167	OpenAI	Proprietary
	GLM-5.2智谱AI	1347.00	+/-7	10,773	智谱AI	Open Source
4	Claude Fable 5Anthropic	1343.00	+/-8.9	6,568	Anthropic	Proprietary
5	Claude Opus 4.6Anthropic	1324.00	+/-4.9	24,030	Anthropic	Proprietary
6	Gemini 3.6 FlashGoogle	1320.00	+/-10.4	4,723	Google	Proprietary
7	Claude Opus 4.6 (thinking)Anthropic	1318.00	+/-5.2	20,665	Anthropic	Proprietary
8	Opus 4.7Anthropic	1318.00	+/-5.4	19,319	Anthropic	Proprietary
9	MiMo-V2.5-ProXiaomi	1313.00	+/-7	10,787	Xiaomi	Open Source
10	Grok 4.5xAI	1313.00	+/-7.1	10,466	xAI	Proprietary
11	GLM 5.1智谱AI	1309.00	+/-6.3	13,604	智谱AI	Open Source
12	Kimi K2.6Moonshot AI	1309.00	+/-4.8	25,390	Moonshot AI	Open Source
13	Qwen3.7 Max阿里巴巴	1309.00	+/-5.5	18,175	阿里巴巴	Proprietary
14	Claude Sonnet 4.6Anthropic	1307.00	+/-5	22,915	Anthropic	Proprietary
15	Muse Spark 1.1Facebook AI研究实验室	1301.00	+/-10.3	4,718	Facebook AI研究实验室	Proprietary
16	GLM-5-Turbo智谱AI	1300.00	+/-4.6	28,063	智谱AI	Proprietary
17	MiMo-V2.5Xiaomi	1292.00	+/-4.4	31,488	Xiaomi	Open Source
18	Kimi K2.7 CodeMoonshot AI	1289.00	+/-7.2	9,920	Moonshot AI	Open Source
19	Gemini 3.5 FlashGoogle Deep Mind	1288.00	+/-5.4	18,576	Google Deep Mind	Proprietary
20	MiniMax M3MiniMaxAI	1285.00	+/-6.1	13,990	MiniMaxAI	Open Source
21	GPT-5.5OpenAI	1283.00	+/-5.6	17,232	OpenAI	Proprietary
22	Muse SparkFacebook AI研究实验室	1282.00	+/-10.9	4,249	Facebook AI研究实验室	Proprietary
23	Qwen3.7-Plus阿里巴巴	1281.00	+/-12.8	2,986	阿里巴巴	Proprietary
24	GLM-5智谱AI	1278.00	+/-3.8	45,606	智谱AI	Open Source
25	DeepSeek-V4-ProDeepSeek-AI	1275.00	+/-5.4	18,497	DeepSeek-AI	Open Source
26	Nex N2 ProNex AGI	1275.00	+/-8.9	6,214	Nex AGI	Open Source
27	Opus 4.5Anthropic	1273.00	+/-4.1	35,317	Anthropic	Proprietary
28	Gemini 3.1 Pro PreviewGoogle Deep Mind	1272.00	+/-4.9	23,843	Google Deep Mind	Proprietary
29	Claude Opus 4.8Anthropic	1270.00	+/-5.5	17,817	Anthropic	Proprietary
30	Kimi K2.5 (thinking)Moonshot AI	1268.00	+/-3.9	40,217	Moonshot AI	Open Source
31	Qwen 3.6 Plus Preview阿里巴巴	1267.00	+/-4.8	24,940	阿里巴巴	Proprietary
32	MiniMax-M2.7MiniMaxAI	1264.00	+/-4.3	32,219	MiniMaxAI	Open Source
33	Hy3腾讯AI实验室	1262.00	+/-19	1,325	腾讯AI实验室	Open Source
34	Gemini 3.1 Pro PreviewGoogle Deep Mind	1260.00	+/-4	36,970	Google Deep Mind	Proprietary
35	GLM-5V-Turbo智谱AI	1259.00	+/-4.3	32,834	智谱AI	Proprietary
36	GPT-5.6 TerraOpenAI	1259.00	+/-12.2	3,312	OpenAI	Proprietary
37	Grok 4.20 Beta (Reasoning)xAI	1249.00	+/-4.5	27,717	xAI	Proprietary
38	GLM-4.7智谱AI	1247.00	+/-3.6	47,089	智谱AI	Open Source
39	GPT-5.6 LunaOpenAI	1247.00	+/-12.6	3,104	OpenAI	Proprietary
40	Hy3腾讯AI实验室	1242.00	+/-12.1	3,352	腾讯AI实验室	Open Source
41	GPT-5.4 (Design Skill, Medium)OpenAI	1242.00	+/-7.2	9,787	OpenAI	Proprietary
42	GPT-5.4 (medium)OpenAI	1240.00	+/-5.3	19,171	OpenAI	Proprietary
43	InklingThinking Machines Lab	1238.00	+/-12.4	3,200	Thinking Machines Lab	Open Source
44	MiniMax M2.5MiniMaxAI	1238.00	+/-6.7	11,504	MiniMaxAI	Open Source
45	DeepSeek-V4-FlashDeepSeek-AI	1236.00	+/-4.6	26,642	DeepSeek-AI	Open Source
46	Grok 4.20 BetaxAI	1230.00	+/-4.4	29,062	xAI	Proprietary
47	M2.1MiniMaxAI	1221.00	+/-5.1	20,805	MiniMaxAI	Open Source
48	Gemini 3.0 FlashGoogle Deep Mind	1220.00	+/-10.6	4,414	Google Deep Mind	Proprietary
49	Grok 4.3 BetaxAI	1218.00	+/-4.9	22,628	xAI	Proprietary
50	Claude Sonnet 4.5Anthropic	1213.00	+/-3.8	42,461	Anthropic	Proprietary
51	Claude Sonnet 4.5 (thinking)Anthropic	1213.00	+/-3.8	41,564	Anthropic	Proprietary
52	GPT-5.4 (low)OpenAI	1210.00	+/-5.1	21,380	OpenAI	Proprietary
53	Qwen3.5-397B-A17B阿里巴巴	1210.00	+/-7.9	8,131	阿里巴巴	Open Source
54	GPT-5.4 (None)OpenAI	1209.00	+/-4.8	23,854	OpenAI	Proprietary
55	GLM-4.7-Flash智谱AI	1208.00	+/-6.6	11,706	智谱AI	Open Source
56	Claude Sonnet 3.7Anthropic	1207.00	+/-5.9	15,245	Anthropic	Proprietary
57	DeepSeek-V3.1 (thinking)DeepSeek-AI	1206.00	+/-5.7	16,258	DeepSeek-AI	Open Source
58	Step 3.7 FlashStepFunAI	1203.00	+/-6	14,567	StepFunAI	Open Source
59	Opus 4.1 (thinking)Anthropic	1202.00	+/-5.8	15,677	Anthropic	Proprietary
60	GPT-5.1 (high)OpenAI	1202.00	+/-5.7	16,056	OpenAI	Proprietary
61	Hy3 (low)腾讯AI实验室	1202.00	+/-11.7	3,721	腾讯AI实验室	Open Source
62	DeepSeek V3.2-ExpDeepSeek-AI	1201.00	+/-5.2	19,490	DeepSeek-AI	Open Source
63	GPT-5.2 (None)OpenAI	1200.00	+/-4.4	29,098	OpenAI	Proprietary
64	GPT-5.2 (medium)OpenAI	1200.00	+/-4.5	28,153	OpenAI	Proprietary
65	GPT-5 (high)OpenAI	1199.00	+/-6.2	13,397	OpenAI	Proprietary
66	Qwen3.5 Plus (0215)阿里巴巴	1198.00	+/-5.3	18,843	阿里巴巴	Proprietary
67	GLM-4.6智谱AI	1197.00	+/-5.6	16,911	智谱AI	Open Source
68	GPT-5.2 (low)OpenAI	1197.00	+/-4.6	25,665	OpenAI	Proprietary
69	GLM-4.5智谱AI	1196.00	+/-5.2	19,637	智谱AI	Open Source
70	GPT-5 (minimal)OpenAI	1196.00	+/-4.2	33,084	OpenAI	Proprietary
71	Opus 4.1Anthropic	1195.00	+/-3.9	39,827	Anthropic	Proprietary
72	DeepSeek V3.2DeepSeek-AI	1195.00	+/-4.4	29,147	DeepSeek-AI	Open Source
73	GPT-5.1 (medium)OpenAI	1192.00	+/-5	21,288	OpenAI	Proprietary
74	Claude Opus 4Anthropic	1191.00	+/-5.6	16,669	Anthropic	Proprietary
75	GPT-5.1 (low)OpenAI	1187.00	+/-4.9	22,157	OpenAI	Proprietary
76	MiMo-V2-FlashXiaomi	1186.00	+/-4	35,604	Xiaomi	Open Source
77	Gemini 2.5 ProGoogle Deep Mind	1184.00	+/-8.6	7,044	Google Deep Mind	Proprietary
78	GPT-5.1 CodexOpenAI	1181.00	+/-16.4	1,807	OpenAI	Proprietary
79	GPT-5.1 (None)OpenAI	1181.00	+/-4.9	22,270	OpenAI	Proprietary
80	GPT-5.2 (high)OpenAI	1180.00	+/-10.8	4,167	OpenAI	Proprietary
81	GPT-5.3 CodexOpenAI	1175.00	+/-5.8	15,668	OpenAI	Proprietary
82	Qwen3-Coder-480B-A35B阿里巴巴	1174.00	+/-16.3	1,958	阿里巴巴	Open Source
83	Mistral Large 3MistralAI	1173.00	+/-4.3	30,690	MistralAI	Open Source
84	Claude Sonnet 4Anthropic	1172.00	+/-5.5	17,532	Anthropic	Proprietary
85	DeepSeek-R1-0528DeepSeek-AI	1169.00	+/-5.4	17,944	DeepSeek-AI	Open Source
86	GLM-4.5-Air智谱AI	1168.00	+/-5.5	17,256	智谱AI	Open Source
87	Claude Sonnet 4 (thinking)Anthropic	1167.00	+/-5.7	16,227	Anthropic	Proprietary
88	MiniMax M2MiniMaxAI	1166.00	+/-6.9	10,828	MiniMaxAI	Open Source
89	AesCoder-4BDesignFlow	1156.00	+/-3.9	39,990	DesignFlow	Open Source
90	Nemotron 3 UltraNVIDIA	1155.00	+/-7.1	11,084	NVIDIA	Open Source
91	Mistral Medium 3.5MistralAI	1153.00	+/-6.9	11,219	MistralAI	Open Source
92	Mistral Medium 3.1 (2508)Mistral	1151.00	+/-4.5	27,888	Mistral	Proprietary
93	Trinity Large ThinkingArcee AI	1147.00	+/-6.4	13,192	Arcee AI	Open Source
94	Haiku 4.5Anthropic	1145.00	+/-4	35,833	Anthropic	Proprietary
95	GPT-5-miniOpenAI	1145.00	+/-4	37,309	OpenAI	Proprietary
96	DeepSeek-V3.1DeepSeek-AI	1143.00	+/-5.1	20,278	DeepSeek-AI	Open Source
97	Qwen3-Max-Thinking阿里巴巴	1140.00	+/-4.2	33,625	阿里巴巴	Proprietary
98	DeepSeek-V3-0324DeepSeek-AI	1139.00	+/-5.2	19,257	DeepSeek-AI	Open Source
99	Prime Intellect: INTELLECT-3Prime Intellect	1137.00	+/-4.3	31,670	Prime Intellect	Open Source
100	Gemini 2.5 Flash-Preview-09-2025Google Deep Mind	1135.00	+/-5.3	19,299	Google Deep Mind	Proprietary
101	Grok 4 FastxAI	1131.00	+/-4	37,026	xAI	Proprietary
102	Kimi K2 0905Moonshot AI	1129.00	+/-17.9	1,504	Moonshot AI	Open Source
103	GPT-5.1 Codex MiniOpenAI	1124.00	+/-4.2	34,100	OpenAI	Proprietary
104	Grok 4.1 FastxAI	1123.00	+/-4.2	33,741	xAI	Proprietary
105	Grok 4.1 Fast (reasoning)xAI	1119.00	+/-4.3	31,458	xAI	Proprietary
106	GPT-5-NanoOpenAI	1115.00	+/-8.6	6,710	OpenAI	Proprietary
107	Kimi K2 Turbo PreviewMoonshot AI	1114.00	+/-15.2	2,094	Moonshot AI	Open Source
108	Gemini 2.5 Flash-Lite-Preview-09-2025Google Deep Mind	1112.00	+/-8.5	6,860	Google Deep Mind	Proprietary
109	Gemini 3.1 Flash-LiteGoogle Deep Mind	1102.00	+/-5	23,904	Google Deep Mind	Proprietary
110	Phi-3-medium 14B-previewMicrosoft Azure	1100.00	+/-8.9	6,396	Microsoft Azure	Proprietary
111	Ministral 3 14BMistralAI	1096.00	+/-14.4	2,379	MistralAI	Open Source
112	Gemini 2.5 FlashGoogle Deep Mind	1090.00	+/-8.5	6,960	Google Deep Mind	Proprietary
113	v0-1.5-mdVercel	1087.00	+/-6.9	11,081	Vercel	Proprietary
114	Ministral 3 8BMistralAI	1084.00	+/-14.3	2,427	MistralAI	Open Source
115	Grok 3xAI	1083.00	+/-4.6	26,862	xAI	Proprietary
116	Grok 4 Fast (reasoning)xAI	1082.00	+/-4.1	37,731	xAI	Proprietary
117	Qwen3-235B-A22B-2507阿里巴巴	1069.00	+/-8.6	6,932	阿里巴巴	Open Source
118	Kimi K2Moonshot AI	1064.00	+/-19.5	1,352	Moonshot AI	Open Source
119	Magistral Medium 1.2 (2509)Mistral	1064.00	+/-9.4	5,851	Mistral	Proprietary
120	Qwen3-235B-A22B-Thinking-2507Alibaba	1063.00	+/-9.1	6,169	Alibaba	Open Source
121	GPT-4.1OpenAI	1056.00	+/-17.3	1,747	OpenAI	Proprietary
122	OpenAI o3OpenAI	1050.00	+/-19.5	1,365	OpenAI	Proprietary
123	Grok 4xAI	1047.00	+/-4.9	23,998	xAI	Proprietary
124	Devstral MediumMistralAI	1043.00	+/-8.6	7,158	MistralAI	Proprietary
125	Ministral 3 3B (2512)Mistral	1041.00	+/-13.5	2,852	Mistral	Open Source
126	Codestral 2508Mistral	1038.00	+/-8.9	6,745	Mistral	Proprietary
127	Qwen3-235B-A22B阿里巴巴	1033.00	+/-10.1	5,154	阿里巴巴	Open Source
128	Grok Code Fast 1xAI	1030.00	+/-11.1	4,295	xAI	Proprietary
129	GPT-4.1 miniOpenAI	1024.00	+/-18.3	1,566	OpenAI	Proprietary
130	Mercury 2Inception	1020.00	+/-9.2	8,763	Inception	Proprietary
131	Magistral Small 1.2 (2509)Mistral	1017.00	+/-9.2	6,448	Mistral	Open Source
132	OpenAI o4 - miniOpenAI	1006.00	+/-16.2	2,011	OpenAI	Proprietary
133	Olmo 3.1 32B ThinkAllen AI	1006.00	+/-6.3	16,162	Allen AI	Open Source
134	GPT OSS 120BOpenAI	994.00	+/-10.3	5,268	OpenAI	Open Source
135	GPT-4.1 nanoOpenAI	993.00	+/-16.9	1,901	OpenAI	Proprietary
136	Qwen3-30B-A3B阿里巴巴	972.00	+/-14.6	2,575	阿里巴巴	Open Source
137	Grok 3 minixAI	961.00	+/-8.8	7,626	xAI	Proprietary
138	Llama 3.1 Nemotron Ultra 253BNVIDIA	960.00	+/-13.8	3,172	NVIDIA	Open Source
139	Mistral-Small-3.2MistralAI	938.00	+/-20.8	1,243	MistralAI	Open Source
140	Llama 4 MaverickFacebook AI研究实验室	910.00	+/-18.4	1,678	Facebook AI研究实验室	Open Source
141	Mistral Large 2.1 (2411)Mistral	894.00	+/-21	1,317	Mistral	Proprietary
142	GPT-4oOpenAI	891.00	+/-18.2	1,780	OpenAI	Proprietary
143	Codestral 2 (2501)Mistral	864.00	+/-20.7	1,444	Mistral	Open Source
144	Devstral Small 1.1MistralAI	838.00	+/-22.5	1,250	MistralAI	Open Source
145	Llama 4 ScoutFacebook AI研究实验室	820.00	+/-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。