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

1394.00

模型数量

163

数据版本

2026年09月08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	1394.00	+/-9.6	5,844	Moonshot AI	Open Source
	GPT-6 Astra (xhigh)OpenAI	1372.00	+/-18.2	1,521	OpenAI	Proprietary
	GPT-5.6 Sol (xhigh)OpenAI	1354.00	+/-8.9	6,448	OpenAI	Proprietary
4	Claude Fable 5.1Anthropic	1346.00	+/-9.5	5,645	Anthropic	Proprietary
5	Claude Opus 5Anthropic	1341.00	+/-6.6	12,061	Anthropic	Proprietary
6	GPT-5.6 Sol (medium)OpenAI	1338.00	+/-5.2	20,271	OpenAI	Proprietary
7	GLM-5.3智谱AI	1334.00	+/-6.6	11,963	智谱AI	Open Source
8	Muse Spark 1.2Facebook AI研究实验室	1331.00	+/-7.1	10,197	Facebook AI研究实验室	Proprietary
9	Claude Fable 5Anthropic	1328.00	+/-5.8	15,916	Anthropic	Proprietary
10	Gemini 3.7 FlashGoogle Deep Mind	1322.00	+/-4.1	36,826	Google Deep Mind	Proprietary
11	Gemini 3.8 FlashGoogle	1322.00	+/-8.8	6,592	Google	Proprietary
12	GLM-5.2智谱AI	1318.00	+/-4.5	27,965	智谱AI	Open Source
13	Grok 4.6xAI	1313.00	+/-4.5	28,686	xAI	Proprietary
14	Qwen3.8 MaxAlibaba	1311.00	+/-9.3	5,698	Alibaba	Open Source
15	Gemini 3.6 FlashGoogle	1308.00	+/-4.8	24,106	Google	Proprietary
16	GLM-5.3-FlashZhipu AI	1308.00	+/-5.8	15,817	Zhipu AI	Open Source
17	Claude Opus 4.6Anthropic	1307.00	+/-4.5	27,379	Anthropic	Proprietary
18	Claude Opus 4.6 (thinking)Anthropic	1303.00	+/-4.8	23,802	Anthropic	Proprietary
19	Deca 2.6 UltraGenLabs	1297.00	+/-16.9	1,670	GenLabs	Proprietary
20	Grok 4.5xAI	1296.00	+/-4.4	29,559	xAI	Proprietary
21	Claude Sonnet 4.6Anthropic	1294.00	+/-4.6	26,295	Anthropic	Proprietary
22	Claude Sonnet 5Anthropic	1294.00	+/-5.5	17,398	Anthropic	Proprietary
23	Muse Spark 1.1Facebook AI研究实验室	1293.00	+/-6.2	13,257	Facebook AI研究实验室	Proprietary
24	Qwen3.7 Max阿里巴巴	1293.00	+/-4.4	29,069	阿里巴巴	Proprietary
25	Kimi K2.6Moonshot AI	1290.00	+/-4.1	33,999	Moonshot AI	Open Source
26	Deca 2.5 MiniGenLabs	1289.00	+/-10.8	4,196	GenLabs	Proprietary
27	MiMo-V2.5-ProXiaomi	1289.00	+/-5.2	19,554	Xiaomi	Open Source
28	GLM 5.1智谱AI	1285.00	+/-4.8	23,577	智谱AI	Open Source
29	GLM-5-Turbo智谱AI	1284.00	+/-4	36,099	智谱AI	Proprietary
30	Qwen3.7-Plus阿里巴巴	1282.00	+/-6.7	11,303	阿里巴巴	Proprietary
31	Deca 2.5 UltraGenLabs	1281.00	+/-11.1	3,917	GenLabs	Proprietary
32	Gemini 3.5 FlashGoogle Deep Mind	1280.00	+/-4.4	29,039	Google Deep Mind	Proprietary
33	MiMo-V2.5Xiaomi	1278.00	+/-3.9	38,899	Xiaomi	Open Source
34	GPT-5.5OpenAI	1276.00	+/-4.5	26,698	OpenAI	Proprietary
35	Kimi K2.7 CodeMoonshot AI	1273.00	+/-5.5	17,077	Moonshot AI	Open Source
36	Muse SparkFacebook AI研究实验室	1269.00	+/-10.9	4,249	Facebook AI研究实验室	Proprietary
37	Claude Opus 4.8Anthropic	1268.00	+/-4.6	25,362	Anthropic	Proprietary
38	MiniMax M3MiniMaxAI	1268.00	+/-4.9	22,330	MiniMaxAI	Open Source
39	GLM-5智谱AI	1264.00	+/-3.6	45,326	智谱AI	Open Source
40	Gemini 3.1 Pro PreviewGoogle Deep Mind	1261.00	+/-4.7	25,342	Google Deep Mind	Proprietary
41	Opus 4.5Anthropic	1260.00	+/-4	35,203	Anthropic	Proprietary
42	DeepSeek-V4-ProDeepSeek-AI	1259.00	+/-4.5	26,466	DeepSeek-AI	Open Source
43	GPT-5.6 TerraOpenAI	1258.00	+/-3.3	67,113	OpenAI	Proprietary
44	Kimi K2.5 (thinking)Moonshot AI	1254.00	+/-3.7	41,601	Moonshot AI	Open Source
45	Qwen 3.6 Plus Preview阿里巴巴	1253.00	+/-4.2	32,471	阿里巴巴	Proprietary
46	MiniMax-M2.7MiniMaxAI	1252.00	+/-3.8	39,539	MiniMaxAI	Open Source
47	GPT-5.6 LunaOpenAI	1250.00	+/-6.1	13,825	OpenAI	Proprietary
48	Nex N2 ProNex AGI	1250.00	+/-6.3	13,001	Nex AGI	Open Source
49	DeepSeek-V4-FlashDeepSeek-AI	1249.00	+/-7.2	9,702	DeepSeek-AI	Open Source
50	Gemini 3.1 Pro PreviewGoogle Deep Mind	1249.00	+/-3.6	45,934	Google Deep Mind	Proprietary
51	GLM-5V-Turbo智谱AI	1247.00	+/-3.8	40,526	智谱AI	Proprietary
52	Hy3腾讯AI实验室	1246.00	+/-5	21,544	腾讯AI实验室	Open Source
53	Grok 4.20 Beta (Reasoning)xAI	1238.00	+/-4	35,817	xAI	Proprietary
54	GLM-4.7智谱AI	1234.00	+/-3.5	46,822	智谱AI	Open Source
55	GPT-5.4 (Design Skill, Medium)OpenAI	1229.00	+/-7.2	9,607	OpenAI	Proprietary
56	GPT-5.4 (medium)OpenAI	1227.00	+/-5.2	18,940	OpenAI	Proprietary
57	InklingThinking Machines Lab	1226.00	+/-12.4	3,184	Thinking Machines Lab	Open Source
58	MiniMax M2.5MiniMaxAI	1225.00	+/-6.7	11,506	MiniMaxAI	Open Source
59	Grok 4.20 BetaxAI	1224.00	+/-3.8	38,926	xAI	Proprietary
60	DeepSeek-V4-FlashDeepSeek-AI	1223.00	+/-4.4	28,074	DeepSeek-AI	Open Source
61	InklingThinking Machines Lab	1217.00	+/-7.9	8,312	Thinking Machines Lab	Open Source
62	M2.1MiniMaxAI	1208.00	+/-5	20,805	MiniMaxAI	Open Source
63	Gemini 3.0 FlashGoogle Deep Mind	1207.00	+/-10.6	4,414	Google Deep Mind	Proprietary
64	Grok 4.3 BetaxAI	1203.00	+/-4.2	32,540	xAI	Proprietary
65	Claude Sonnet 4.5Anthropic	1200.00	+/-3.5	45,423	Anthropic	Proprietary
66	Claude Sonnet 4.5 (thinking)Anthropic	1200.00	+/-3.6	44,182	Anthropic	Proprietary
67	Step 3.7 FlashStepFunAI	1198.00	+/-4.9	22,607	StepFunAI	Open Source
68	GPT-5.4 (low)OpenAI	1197.00	+/-5	21,133	OpenAI	Proprietary
69	Qwen3.5-397B-A17B阿里巴巴	1197.00	+/-7.8	8,129	阿里巴巴	Open Source
70	GPT-5.4OpenAI	1196.00	+/-4.7	23,597	OpenAI	Proprietary
71	GLM-4.7-Flash智谱AI	1195.00	+/-6.6	11,706	智谱AI	Open Source
72	Claude Sonnet 3.7Anthropic	1194.00	+/-5.8	15,259	Anthropic	Proprietary
73	Hy3 (low)腾讯AI实验室	1194.00	+/-11.1	4,063	腾讯AI实验室	Open Source
74	DeepSeek-V3.1 (thinking)DeepSeek-AI	1193.00	+/-5.6	16,257	DeepSeek-AI	Open Source
75	Opus 4.1 (thinking)Anthropic	1188.00	+/-5.7	15,715	Anthropic	Proprietary
76	DeepSeek V3.2-ExpDeepSeek-AI	1188.00	+/-5.2	19,487	DeepSeek-AI	Open Source
77	GPT-5.1 (high)OpenAI	1188.00	+/-5.6	16,116	OpenAI	Proprietary
78	GPT-5.2 (medium)OpenAI	1187.00	+/-4.4	27,932	OpenAI	Proprietary
79	GPT-5.2 (None)OpenAI	1186.00	+/-4.3	28,872	OpenAI	Proprietary
80	GPT-5 (high)OpenAI	1185.00	+/-6.1	13,485	OpenAI	Proprietary
81	Qwen3.5 Plus (0215)阿里巴巴	1184.00	+/-5.3	18,498	阿里巴巴	Proprietary
82	Opus 4.1Anthropic	1183.00	+/-3.7	39,733	Anthropic	Proprietary
83	GLM-4.5智谱AI	1183.00	+/-5.1	19,653	智谱AI	Open Source
84	GLM-4.6智谱AI	1183.00	+/-5.5	17,086	智谱AI	Open Source
85	GPT-5 (minimal)OpenAI	1183.00	+/-4.1	32,965	OpenAI	Proprietary
86	GPT-5.2 (low)OpenAI	1183.00	+/-4.6	25,456	OpenAI	Proprietary
87	DeepSeek V3.2DeepSeek-AI	1182.00	+/-4.3	29,120	DeepSeek-AI	Open Source
88	GPT-5.1 (medium)OpenAI	1179.00	+/-4.9	21,347	OpenAI	Proprietary
89	Claude Opus 4Anthropic	1178.00	+/-5.6	16,668	Anthropic	Proprietary
90	MiMo-V2-FlashXiaomi	1174.00	+/-4	35,226	Xiaomi	Open Source
91	GPT-5.1 (low)OpenAI	1173.00	+/-4.9	22,192	OpenAI	Proprietary
92	Solar Pro 4Upstage	1171.00	+/-13.9	2,692	Upstage	Proprietary
93	Gemini 2.5 ProGoogle Deep Mind	1170.00	+/-8.4	7,220	Google Deep Mind	Proprietary
94	GPT-5.1 CodexOpenAI	1170.00	+/-15.5	2,005	OpenAI	Proprietary
95	GPT-5.1 (None)OpenAI	1168.00	+/-4.8	22,295	OpenAI	Proprietary
96	GPT-5.2 (high)OpenAI	1168.00	+/-10.8	4,150	OpenAI	Proprietary
97	GPT-5.3 CodexOpenAI	1162.00	+/-5.7	15,673	OpenAI	Proprietary
98	v0-1.5-lgVercel	1161.00	+/-21.4	1,192	Vercel	Proprietary
99	Mistral Large 3MistralAI	1160.00	+/-4.2	30,428	MistralAI	Open Source
100	Qwen3-Coder-480B-A35B阿里巴巴	1160.00	+/-16.3	1,955	阿里巴巴	Open Source
101	Claude Sonnet 4Anthropic	1159.00	+/-5.4	17,542	Anthropic	Proprietary
102	DeepSeek-R1-0528DeepSeek-AI	1156.00	+/-5.4	17,951	DeepSeek-AI	Open Source
103	GLM-4.5-Air智谱AI	1155.00	+/-5.5	17,253	智谱AI	Open Source
104	AGI-01 SwiftLucidQuery	1154.00	+/-20.2	1,248	LucidQuery	Proprietary
105	Claude Sonnet 4 (thinking)Anthropic	1154.00	+/-5.6	16,246	Anthropic	Proprietary
106	Nemotron 3 UltraNVIDIA	1154.00	+/-5.5	18,552	NVIDIA	Open Source
107	MiniMax M2MiniMaxAI	1153.00	+/-6.8	10,828	MiniMaxAI	Open Source
108	AesCoder-4BDesignFlow	1142.00	+/-3.8	39,833	DesignFlow	Open Source
109	Mistral Medium 3.5MistralAI	1139.00	+/-6.9	10,892	MistralAI	Open Source
110	Mistral Medium 3.1 (2508)Mistral	1138.00	+/-4.4	27,823	Mistral	Proprietary
111	Trinity Large ThinkingArcee AI	1135.00	+/-6.4	12,837	Arcee AI	Open Source
112	Haiku 4.5Anthropic	1132.00	+/-4	35,537	Anthropic	Proprietary
113	GPT-5-miniOpenAI	1132.00	+/-3.9	37,081	OpenAI	Proprietary
114	DeepSeek-V3.1DeepSeek-AI	1129.00	+/-5.1	20,334	DeepSeek-AI	Open Source
115	Qwen3-Max-Thinking阿里巴巴	1127.00	+/-4.1	33,365	阿里巴巴	Proprietary
116	DeepSeek-V3-0324DeepSeek-AI	1126.00	+/-5.2	19,271	DeepSeek-AI	Open Source
117	Prime Intellect: INTELLECT-3Prime Intellect	1124.00	+/-4.2	31,341	Prime Intellect	Open Source
118	Gemini 2.5 Flash-Preview-09-2025Google Deep Mind	1122.00	+/-5.2	19,295	Google Deep Mind	Proprietary
119	Grok 4 FastxAI	1118.00	+/-3.9	36,645	xAI	Proprietary
120	Kimi K2 0905Moonshot AI	1115.00	+/-17.9	1,504	Moonshot AI	Open Source
121	GPT-5.1 Codex MiniOpenAI	1111.00	+/-4.1	33,827	OpenAI	Proprietary
122	Grok 4.1 FastxAI	1110.00	+/-4.1	33,597	xAI	Proprietary
123	Grok 4.1 Fast (reasoning)xAI	1104.00	+/-4.2	31,131	xAI	Proprietary
124	GPT-5-NanoOpenAI	1102.00	+/-8.6	6,703	OpenAI	Proprietary
125	Kimi K2 Turbo PreviewMoonshot AI	1101.00	+/-15.2	2,094	Moonshot AI	Open Source
126	Gemini 2.5 Flash-Lite-Preview-09-2025Google Deep Mind	1099.00	+/-8.5	6,850	Google Deep Mind	Proprietary
127	Gemini 3.1 Flash-LiteGoogle Deep Mind	1090.00	+/-4.9	23,527	Google Deep Mind	Proprietary
128	Phi-3-medium 14B-previewMicrosoft Azure	1086.00	+/-8.9	6,390	Microsoft Azure	Proprietary
129	Ministral 3 14BMistralAI	1082.00	+/-14.4	2,380	MistralAI	Open Source
130	Gemini 2.5 FlashGoogle Deep Mind	1077.00	+/-8.5	6,971	Google Deep Mind	Proprietary
131	v0-1.5-mdVercel	1074.00	+/-6.8	11,081	Vercel	Proprietary
132	Ministral 3 8BMistralAI	1071.00	+/-14.3	2,427	MistralAI	Open Source
133	Grok 3xAI	1070.00	+/-4.5	26,865	xAI	Proprietary
134	Grok 4 Fast (reasoning)xAI	1068.00	+/-4	37,369	xAI	Proprietary
135	Qwen3-235B-A22B-2507阿里巴巴	1056.00	+/-8.6	6,939	阿里巴巴	Open Source
136	Kimi K2Moonshot AI	1051.00	+/-19.5	1,352	Moonshot AI	Open Source
137	Magistral Medium 1.2 (2509)Mistral	1050.00	+/-9.3	5,854	Mistral	Proprietary
138	Qwen3-235B-A22B-Thinking-2507Alibaba	1050.00	+/-9.1	6,175	Alibaba	Open Source
139	GPT-4.1OpenAI	1043.00	+/-17.3	1,747	OpenAI	Proprietary
140	OpenAI o3OpenAI	1037.00	+/-19.5	1,365	OpenAI	Proprietary
141	Grok 4xAI	1034.00	+/-4.9	24,016	xAI	Proprietary
142	Devstral MediumMistralAI	1030.00	+/-8.5	7,153	MistralAI	Proprietary
143	Ministral 3 3B (2512)Mistral	1028.00	+/-13.5	2,852	Mistral	Open Source
144	Mercury 2Inception	1026.00	+/-6.6	17,309	Inception	Proprietary
145	Codestral 2508Mistral	1025.00	+/-8.8	6,734	Mistral	Proprietary
146	Qwen3-235B-A22B阿里巴巴	1019.00	+/-10.1	5,158	阿里巴巴	Open Source
147	Grok Code Fast 1xAI	1016.00	+/-11.1	4,299	xAI	Proprietary
148	GPT-4.1 miniOpenAI	1011.00	+/-18.3	1,566	OpenAI	Proprietary
149	Magistral Small 1.2 (2509)Mistral	1003.00	+/-9.2	6,447	Mistral	Open Source
150	OpenAI o4 - miniOpenAI	993.00	+/-16.2	2,011	OpenAI	Proprietary
151	Olmo 3.1 32B ThinkAllen AI	993.00	+/-6.2	16,164	Allen AI	Open Source
152	GPT-4.1 nanoOpenAI	980.00	+/-16.9	1,901	OpenAI	Proprietary
153	GPT OSS 120BOpenAI	980.00	+/-10.3	5,272	OpenAI	Open Source
154	Qwen3-30B-A3B阿里巴巴	959.00	+/-14.6	2,568	阿里巴巴	Open Source
155	Grok 3 minixAI	948.00	+/-8.8	7,621	xAI	Proprietary
156	Llama 3.1 Nemotron Ultra 253BNVIDIA	947.00	+/-13.8	3,173	NVIDIA	Open Source
157	Mistral-Small-3.2MistralAI	924.00	+/-20.8	1,243	MistralAI	Open Source
158	Llama 4 MaverickFacebook AI研究实验室	897.00	+/-18.4	1,678	Facebook AI研究实验室	Open Source
159	Mistral Large 2.1 (2411)Mistral	881.00	+/-21	1,317	Mistral	Proprietary
160	GPT-4oOpenAI	878.00	+/-18.2	1,780	OpenAI	Proprietary
161	Codestral 2 (2501)Mistral	851.00	+/-20.7	1,444	Mistral	Open Source
162	Devstral Small 1.1MistralAI	825.00	+/-22.6	1,250	MistralAI	Open Source
163	Llama 4 ScoutFacebook AI研究实验室	807.00	+/-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。