🔭

AI前端开发排行榜 - 大模型写网页代码能力排名

Arena

AI大模型前端开发能力排行榜，基于LMSYS Arena盲测，测试模型编写React、HTML、小游戏、数据面板等前端代码的能力，涵盖Claude、GPT、DeepSeek等编程强模型

总榜排名↑	模型名称	研发机构	React	HTML	小游戏	数据面板	品牌页	消费产品	设计还原	创作工具	物理模拟
1	Claude 4.7 Opus Thinking	Anthropic	1	1	1	2	1	1	1	1	2
2	Claude 4.7 Opus	Anthropic	2	2	2	1	2	2	2	3	1
3	Claude 4.6 Opus Thinking	Anthropic	3	4	3	6	4	4	3	4	4
4	Claude 4.6 Opus	Anthropic	4	3	4	3	5	3	4	2	5
5	GLM 5.1	Zhipu	5	5	5	5	8	6	5	6	3
6	Claude 4.6 Sonnet	Anthropic	7	7	7	4	7	5	6	5	8
7	Kimi K2.6	moonshot	6	9	11	7	3	7	8	8	7
8	Muse Spark	Meta	8	6	10	10	6	9	9	10	11
9	GPT 5.5 High	OpenAI	9	8	6	11	11	14	10	7	6
10	Claude 4.5 Opus 20251101 Thinking 32K	Anthropic	10	10	8	8	9	8	7	9	12
11	Qwen 3.6 Max Preview	alibaba	11	15	13	9	10	10	11	12	10
12	MiMo V2.5 Pro	Xiaomi	12	13	12	14	14	15	13	13	13
13	Claude 4.5 Opus 20251101	Anthropic	13	14	16	15	16	12	14	17	17
14	Qwen 3.6 Plus	alibaba	14	18	18	13	15	11	15	11	14
15	GPT 5.4 High	OpenAI	15	20	9	26	18	18	18	15	9
17	Gemini 3.1 Pro Preview	Google	17	11	14	25	13	24	16	22	15
18	MiMo V2.5	xiaomi	18	22	21	17	19	17	26	19	33
19	GPT 5.5	OpenAI	21	12	17	21	26	25	20	23	18
20	GLM 4.7	Zhipu	19	23	25	16	20	16	21	21	29
21	Gemini 3 Pro	Google	26	19	20	20	17	22	19	27	20
22	GPT 5.4 Medium	OpenAI	22	16	15	22	23	26	17	14	16
24	GLM 5	Zhipu	23	21	24	18	24	19	22	16	21
25	MiMo V2 Pro	Xiaomi	24	25	29	19	27	23	24	24	19
26	Kimi K2.5 Thinking	Moonshot	25	30	27	23	21	20	23	26	24
27	Kimi K2.5 Instant	Moonshot	30	31	34	40	25	32	27	36	38
30	GPT 5.2 Chat	OpenAI	—	32	22	28	38	34	28	39	43
31	GPT 5.4 Mini High	OpenAI	32	26	28	29	37	39	30	33	23
32	Grok 4.3	xAI	28	50	41	34	28	41	37	42	32
33	Grok 4.20 Beta Reasoning	xAI	31	43	38	33	31	35	36	43	42
34	GPT 5 Medium	OpenAI	—	35	30	46	29	29	29	47	52
36	GPT 5.1 Medium	OpenAI	—	38	33	32	34	31	32	37	47
37	Gemini 3 Flash	Google	37	36	36	41	33	38	35	40	34
38	Claude 4.5 Sonnet 20250929 Thinking 32K	Anthropic	34	39	37	31	41	33	38	38	35
39	Qwen 3.5 397B A17B	alibaba	36	37	40	39	36	40	40	35	36
40	Claude 4.5 Sonnet 20250929	Anthropic	33	41	39	36	42	37	39	34	37
41	GPT 5.4	OpenAI	42	—	—	—	—	—	—	—	—
42	Claude 4.1 Opus 20250805	Anthropic	—	40	31	42	45	43	33	44	41
43	Gemma 4 31B	Google	35	49	51	48	39	30	47	46	30
45	GPT 5.3 Codex	OpenAI	43	29	32	44	48	49	41	32	31
47	Hunyuan HY3 Preview	Tencent	41	47	43	52	47	47	43	29	40
48	Qwen 3.5 122B A10B	alibaba	44	53	46	45	44	45	45	45	46
49	Gemma 4 26B A4B	Google	45	51	45	54	35	52	46	54	54
50	GLM 4.6	Zhipu	—	45	49	47	52	46	50	50	45
51	Qwen 3.5 27B	alibaba	47	42	52	43	49	48	49	48	48
54	GPT 5.2 Codex	OpenAI	48	54	54	51	51	51	51	49	49
56	Kimi K2 Thinking Turbo	Moonshot	50	56	55	56	55	56	56	53	51
57	GPT 5.1 Codex	OpenAI	—	55	48	50	50	53	53	57	56
58	Claude 4.5 Haiku 20251001	Anthropic	49	59	57	55	59	57	57	55	57
60	MiMo V2 Flash	Xiaomi	54	52	58	60	57	59	59	60	58
62	Qwen 3 Coder 480B A35B Instruct	alibaba	53	63	62	61	60	61	60	59	60
63	AI Kat Coder Pro V1	Unknown	—	66	66	59	62	62	63	63	66
64	Qwen 3.5 35B A3B	alibaba	56	57	68	63	63	63	62	66	64
65	trinity-large-thinking	Arcee	55	64	61	67	65	64	66	70	63
66	Gemini 3.1 Flash Lite Preview	Google	57	65	63	71	69	67	64	65	62
67	GPT 5.1 Codex Mini	OpenAI	—	67	64	64	68	65	68	62	68
68	Qwen 3.5 Flash	alibaba	59	61	70	68	64	69	65	67	70
69	Grok 4.1 Fast Reasoning	xAI	58	68	65	70	66	66	67	64	67
71	Grok 4.1 Thinking	xAI	—	70	67	66	72	71	70	68	74
72	Gemini 2.5 Pro	Google	—	72	71	73	70	72	73	71	69
73	AI Devstral 2	Unknown	60	71	73	69	71	68	71	69	71
74	Mercury 2	Inception Labs	61	73	72	74	76	75	72	—	—
75	Grok 4 Fast Reasoning	xAI	—	74	74	72	73	74	74	74	73
76	Grok Code Fast 1	xAI	—	75	75	75	74	73	75	73	72
77	AI Devstral Medium 2507	Unknown	—	76	76	76	75	76	76	75	75