Code category rankings

Best AI for code categories

Compare models on Design Arena code-category tournaments. These results stay separate from the main Vibe Coding Index and the Arena Code WebDev board.

Current source snapshot

Jul 19, 2026

132 ranked entries

Live category evidence

Current Design Arena leaderboard

How this source works →

Rank	Model	ELO	Win rate	Average time	Blended price	Tournament runs
1	GLM 5.2 Zhipu	1,345	61.2%	337.8s	$2.15	11,609
2	Claude Fable 5 Anthropic	1,343	62.8%	151.7s	$22	9,212
3	Claude Opus 4.6 Anthropic	1,330	63.3%	187.9s	$11	19,768
4	Grok 4.5 xAI	1,328	58.2%	125.8s	$3	8,775
5	Claude Opus 4.6 (Thinking) Anthropic	1,325	62.5%	189.3s	$11	11,891
6	Claude Opus 4.7 Anthropic	1,325	60.4%	141.5s	$11	19,576
7	GLM 5.1 Zhipu	1,316	58.9%	314.4s	$2.14	6,832
8	Kimi K2.6 Moonshot	1,314	57.9%	414.0s	$1.71	28,760
9	Claude Sonnet 4.6 Anthropic	1,313	60.8%	172.0s	$6	20,333
10	Claude Sonnet 5 Anthropic	1,312	56.8%	143.8s	$4	5,230
11	Muse Spark 1.1 Meta	1,307	55.6%	193.1s	$2	13,299
12	GLM 5 Turbo Zhipu	1,306	57.1%	330.4s	$1.9	12,894
13	MiMo-V2.5-Pro Xiaomi	1,305	57.1%	297.1s	$1.5	5,091
14	Qwen3.7 Max Alibaba	1,302	57.1%	255.7s	$2.21	13,642
15	Kimi K2.7 Code Moonshot	1,294	54.5%	308.8s	$1.71	11,579
16	Gemini 3.5 Flash Google	1,292	56.9%	72.6s	$3.38	29,829
17	MiniMax M3 MiniMax	1,290	55.0%	185.5s	$0.53	7,461
18	Qwen3.7 Plus Alibaba	1,290	52.6%	318.2s	$0.56	956
19	MiMo-V2.5 Xiaomi	1,289	54.6%	176.5s	$0.14	14,558
20	GPT-5.5 OpenAI	1,288	56.4%	149.5s	$11.25	23,121
21	GLM 5 Zhipu	1,283	55.5%	197.5s	$1.55	32,017
22	Claude Opus 4.5 Anthropic	1,279	59.6%	124.2s	$10	18,998
23	Gemini 3 Pro Preview Google	1,278	64.2%	88.7s	$4.5	12,275
24	DeepSeek-V4-Pro DeepSeek	1,275	54.3%	308.4s	$2.18	10,272
25	Claude Opus 4.8 Anthropic	1,273	54.3%	85.8s	$10	16,311
26	Kimi K2.5 (Thinking) Moonshot	1,273	54.5%	218.1s	$1.3	27,147
27	MiniMax M2.7 MiniMax	1,269	53.3%	190.6s	$0.53	14,675
28	Gemini 3.1 Pro Preview Google	1,266	56.8%	99.3s	$4.5	36,401
29	GLM 5V Turbo Zhipu	1,265	52.3%	181.3s	$1.9	29,832
30	Qwen3.6 Plus Alibaba	1,260	51.4%	260.3s	$0.73	16,566
31	Nex N2 Pro Nex Agi	1,258	48.9%	453.6s	$0.44	3,755
32	Grok 4.20 Beta (Reasoning) xAI	1,255	54.3%	94.0s	$1.56	17,038
33	GLM 4.7 Zhipu	1,253	54.7%	166.0s	$1	29,268
34	GPT-5.4 (Design Skill, Medium) OpenAI	1,248	52.6%	163.4s	$5.63	4,833
35	GPT-5.4 (Medium) OpenAI	1,246	52.8%	165.6s	$5.63	10,752
36	MiniMax M2.5 MiniMax	1,244	56.8%	137.5s	$0.44	5,389
37	DeepSeek-V4-Flash DeepSeek	1,242	49.7%	130.5s	$0.17	11,380
38	Hy3 Tencent	1,238	46.1%	256.0s	$0.25	1,047
39	Grok 4.20 Beta xAI	1,236	51.9%	68.3s	$1.56	17,353
40	MiniMax M2.1 MiniMax	1,227	55.3%	120.4s	$0.53	9,381
41	Claude 3.7 Sonnet Anthropic	1,226	57.6%	117.0s	—	7,529
42	Gemini 3 Flash Preview Google	1,226	57.6%	34.5s	$1.13	9,357
43	Grok 4.3 xAI	1,226	48.4%	83.5s	$1.56	29,451
44	Claude Sonnet 4.5 Anthropic	1,219	52.5%	97.3s	$6.6	21,364
45	Claude Sonnet 4.5 (Thinking) Anthropic	1,219	52.4%	92.4s	$6.6	20,943
46	GPT-5.4 (Low) OpenAI	1,216	49.3%	86.0s	$5.63	11,777
47	Qwen3.5 397B A17B Alibaba	1,216	52.6%	135.7s	$1.09	4,149
48	GPT-5.4 (None) OpenAI	1,215	49.3%	75.4s	$5.63	13,539
49	GLM 4.7 Flash Zhipu	1,214	53.1%	150.1s	$0.15	5,798
50	DeepSeek-V3.1 (Thinking) DeepSeek	1,212	56.0%	197.5s	$0.44	7,784
51	Claude Opus 4.1 (Thinking) Anthropic	1,208	55.8%	104.5s	$30	7,473
52	GPT-5.1 (High) OpenAI	1,208	53.1%	165.8s	$3.44	8,405
53	DeepSeek-V3.2-Exp DeepSeek	1,207	54.2%	197.7s	$0.31	9,583
54	GPT-5.2 (None) OpenAI	1,206	49.8%	167.3s	$4.81	16,148
55	GPT-5.2 (Medium) OpenAI	1,206	49.6%	173.6s	$4.81	15,472
56	GPT-5 (High) OpenAI	1,205	54.7%	164.6s	$3.44	6,807
57	MiMo-V2-Flash Xiaomi	1,205	49.4%	62.2s	—	16,624
58	Qwen3.5 Plus 02-15 Alibaba	1,204	48.5%	131.4s	$0.59	9,623
59	Step 3.7 Flash Stepfun	1,204	44.8%	102.0s	$0.44	8,234
60	GLM 4.6 Zhipu	1,203	54.3%	109.5s	$0.96	14,578
61	GPT-5.2 (Low) OpenAI	1,203	50.3%	166.3s	$4.81	13,533
62	GLM 4.5 Zhipu	1,202	54.4%	141.4s	$1	9,161
63	GPT-5 (Minimal) OpenAI	1,202	51.9%	120.2s	$3.44	17,333
64	Claude Opus 4.1 Anthropic	1,201	50.1%	105.6s	$30	20,472
65	DeepSeek-V3.2 DeepSeek	1,200	49.3%	183.2s	$1	14,620
66	GPT-5.1 (Medium) OpenAI	1,198	52.0%	135.6s	$3.44	10,709
67	Claude Opus 4 Anthropic	1,197	55.6%	88.5s	$30	7,929
68	GPT-5.1 (Low) OpenAI	1,193	51.5%	113.2s	$3.44	10,832
69	Gemini 2.5 Pro Google	1,191	58.3%	74.3s	$3.44	3,849
70	Hy3 Low Tencent	1,188	39.2%	85.8s	$0.25	1,268
71	GPT-5.1 Codex OpenAI	1,187	55.2%	74.0s	$3.44	1,931
72	GPT-5.1 (None) OpenAI	1,187	50.7%	107.8s	$3.44	10,899
73	GPT-5.2 (High) OpenAI	1,186	50.4%	152.0s	$4.81	1,661
74	GPT-5.3 Codex OpenAI	1,181	47.3%	52.4s	$4.81	11,788
75	Qwen3 Coder 480B A35B Instruct Alibaba	1,180	61.2%	62.1s	$0.67	902
76	Claude Sonnet 4 Anthropic	1,178	53.4%	72.9s	$6	8,316
77	Mistral Large 3 (2512) Mistral	1,178	47.6%	110.0s	$0.75	14,978
78	DeepSeek-R1-0528 DeepSeek	1,175	52.6%	88.6s	$1.15	8,357
79	GLM 4.5 Air Zhipu	1,174	51.5%	121.1s	$0.43	8,103
80	Claude Sonnet 4 (Thinking) Anthropic	1,173	51.6%	71.4s	$6	7,680
81	MiniMax M2 Stable MiniMax	1,172	48.1%	121.2s	$0.45	5,275
82	Nemotron 3 Ultra NVIDIA	1,159	36.5%	139.9s	$1.05	4,921
83	Mistral Medium 3.1 (2508) Mistral	1,157	45.1%	114.2s	$0.8	14,173
84	Prime Intellect. INTELLECT-3 Prime Intellect	1,156	42.6%	78.9s	—	15,077
85	Trinity Large Thinking Arcee Ai	1,152	40.1%	89.3s	$0.39	6,482
86	Claude Haiku 4.5 Anthropic	1,151	44.8%	48.3s	$2.2	18,482
87	GPT-5 mini (Default) OpenAI	1,151	43.4%	128.6s	$0.69	19,372
88	Grok 4 Fast xAI	1,150	43.0%	45.5s	—	18,019
89	DeepSeek-V3.1 DeepSeek	1,149	47.9%	45.6s	$0.46	9,535
90	Qwen3 Max Alibaba	1,146	44.0%	108.3s	$1.56	16,354
91	DeepSeek-V3-0324 DeepSeek	1,145	48.5%	74.0s	$0.63	9,090
92	Grok 4.1 Fast xAI	1,142	42.1%	34.8s	—	16,318
93	Gemini 2.5 Flash Preview 09-2025 Google	1,141	46.9%	45.7s	$0.85	9,097
94	Grok 4.1 Fast (Reasoning) xAI	1,137	41.3%	64.8s	—	16,053
95	Kimi K2 0905 Preview Moonshot	1,135	48.5%	194.1s	$1.08	672
96	GPT-5.1 Codex Mini OpenAI	1,130	41.5%	47.5s	$0.69	17,161
97	GPT-5 nano (Default) OpenAI	1,121	48.1%	90.8s	$0.14	3,043
98	Gemini 2.5 Flash Lite Preview 09-2025 Google	1,118	47.0%	13.3s	—	3,139
99	Gemini 3.1 Flash-Lite Preview Google	1,108	36.4%	5.8s	$0.56	12,911
100	Mistral Medium 3 (2505) Mistral	1,106	48.1%	74.2s	$0.8	2,890
101	Grok 3 xAI	1,102	40.1%	45.1s	—	13,192
102	Ministral 3 14B (2512) Mistral	1,102	44.0%	81.9s	$0.35	1,017
103	Grok 4 Fast (Reasoning) xAI	1,100	36.8%	26.6s	—	18,584
104	Gemini 2.5 Flash Google	1,096	45.4%	74.5s	$0.85	3,254
105	Ministral 3 8B (2512) Mistral	1,090	42.9%	52.9s	$0.3	1,034
106	Qwen3-235B-A22B-Instruct-2507 Alibaba	1,075	42.7%	53.8s	$0.3	3,206
107	Kimi K2 Moonshot	1,070	51.7%	60.0s	$1	596
108	Qwen3-235B-A22B-Thinking-2507 Alibaba	1,070	40.9%	112.6s	$0.75	2,908
109	Grok 4 xAI	1,065	35.1%	63.7s	—	11,878
110	Devstral Medium Mistral	1,062	41.2%	24.7s	—	3,285
111	GPT-4.1 OpenAI	1,062	50.9%	72.7s	$3.5	763
112	o3 OpenAI	1,057	51.9%	52.2s	$3.5	590
113	Grok Code Fast 1 xAI	1,048	36.9%	16.7s	—	8,460
114	Ministral 3 3B (2512) Mistral	1,047	37.3%	36.3s	$0.15	1,253
115	Codestral 2508 Mistral	1,044	38.5%	17.5s	$0.45	3,142
116	Qwen3-235B-A22B Alibaba	1,039	38.3%	78.1s	$0.8	2,399
117	GPT-4.1 mini OpenAI	1,030	47.5%	71.4s	$0.7	688
118	Mercury 2 Inception	1,028	21.3%	4.3s	$0.38	3,943
119	Olmo 3.1 32B Think Allenai	1,024	27.1%	128.1s	—	8,236
120	o4-mini OpenAI	1,013	46.4%	38.9s	$1.93	885
121	GPT-4.1 nano OpenAI	1,000	47.3%	24.2s	$0.18	858
122	GPT OSS 120B OpenAI	1,000	33.4%	24.1s	$0.26	2,487
123	Grok 3 Mini xAI	980	32.4%	23.9s	—	3,594
124	Qwen3 30B-A3B Alibaba	979	37.5%	28.8s	$0.24	1,184
125	Llama 3.1 Nemotron Ultra 253B NVIDIA	978	27.2%	32.2s	—	1,464
126	Mistral Small 3.2 Mistral	944	39.8%	17.0s	$0.14	567
127	Llama 4 Maverick Meta Llama	916	35.8%	13.1s	$0.35	790
128	Mistral Large 2.1 (2411) Mistral	912	33.3%	32.6s	—	584
129	GPT-4o OpenAI	897	34.8%	14.2s	$4.38	850
130	Codestral 2 (2501) Mistral	870	30.2%	6.1s	$0.45	702
131	Devstral Small 1.1 Mistral	856	27.7%	6.9s	—	600
132	Llama 4 Scout Meta Llama	826	26.6%	12.0s	$0.15	615

What this ranking means. ELO summarizes head-to-head preference results inside this category. It is useful evidence for the named task, but it remains separate from the broader Vibe Coding Index.