Game development rankings

Best AI for game development

Compare models on game-building work using current Design Arena preference results. The ranking stays separate from the main Vibe Coding Index.

Current source snapshot

Jul 17, 2026

129 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	Claude Fable 5 Anthropic	1,396	66.7%	175.9s	$22	8,663
2	GLM 5.2 Zhipu	1,349	60.8%	411.4s	$2.15	11,176
3	Claude Sonnet 5 Anthropic	1,345	59.9%	145.0s	$4	5,230
4	GPT-5.5 OpenAI	1,343	62.2%	161.7s	$11.25	22,204
5	Claude Opus 4.6 (Thinking) Anthropic	1,337	63.0%	185.2s	$11	11,862
6	Claude Opus 4.7 Anthropic	1,336	61.2%	147.5s	$11	19,389
7	Claude Opus 4.6 Anthropic	1,335	62.9%	193.1s	$11	19,598
8	MiMo-V2.5-Pro Xiaomi	1,331	61.0%	296.6s	$1.5	5,024
9	Grok 4.5 xAI	1,331	54.5%	110.6s	$3	5,992
10	GLM 5.1 Zhipu	1,328	61.4%	404.0s	$2.14	6,779
11	Muse Spark 1.1 Meta	1,327	56.1%	248.4s	$2	10,843
12	Gemini 3.5 Flash Google	1,319	57.6%	86.6s	$3.38	28,872
13	Claude Sonnet 4.6 Anthropic	1,315	60.2%	177.0s	$6	20,153
14	GLM 5 Turbo Zhipu	1,315	59.0%	402.1s	$1.9	12,847
15	Qwen3.7 Max Alibaba	1,315	57.9%	243.1s	$2.21	12,830
16	Kimi K2.6 Moonshot	1,307	57.5%	420.5s	$1.71	28,120
17	Claude Opus 4.8 Anthropic	1,302	55.1%	101.9s	$10	15,466
18	GPT-5.4 (Medium) OpenAI	1,298	57.8%	202.5s	$5.63	10,752
19	GLM 5 Zhipu	1,292	57.4%	190.3s	$1.55	31,968
20	MiMo-V2.5 Xiaomi	1,291	55.4%	188.0s	$0.14	14,447
21	Claude Opus 4.5 Anthropic	1,288	59.4%	126.1s	$10	18,995
22	DeepSeek-V4-Pro DeepSeek	1,288	55.8%	375.6s	$2.18	10,218
23	GLM 5V Turbo Zhipu	1,280	54.5%	163.8s	$1.9	29,673
24	MiniMax M3 MiniMax	1,280	51.5%	129.4s	$0.53	7,186
25	Qwen3.7 Plus Alibaba	1,276	48.3%	327.9s	$0.56	848
26	Kimi K2.7 Code Moonshot	1,270	50.6%	304.7s	$1.71	10,711
27	MiniMax M2.7 MiniMax	1,265	53.0%	207.5s	$0.53	14,613
28	Kimi K2.5 (Thinking) Moonshot	1,264	53.5%	229.4s	$1.3	27,045
29	Qwen3.6 Plus Alibaba	1,262	51.8%	269.0s	$0.73	16,468
30	Nex N2 Pro Nex Agi	1,261	48.5%	484.2s	$0.44	3,681
31	Gemini 3.1 Pro Preview Google	1,257	54.3%	102.1s	$4.5	36,203
32	Grok 4.20 Beta (Reasoning) xAI	1,255	53.5%	93.1s	$1.56	16,956
33	DeepSeek-V4-Flash DeepSeek	1,254	50.4%	172.6s	$0.17	11,320
34	GPT-5.2 (Low) OpenAI	1,253	56.0%	181.1s	$4.81	13,533
35	Gemini 3 Pro Preview Google	1,253	60.6%	95.5s	$4.5	12,275
36	GPT-5.4 (None) OpenAI	1,252	53.6%	91.4s	$5.63	13,537
37	GPT-5.4 (Low) OpenAI	1,250	52.9%	95.4s	$5.63	11,777
38	GLM 4.7 Zhipu	1,247	55.1%	167.7s	$1	29,267
39	GPT-5 (High) OpenAI	1,245	59.5%	175.6s	$3.44	6,807
40	GPT-5.2 (High) OpenAI	1,245	56.7%	169.8s	$4.81	1,661
41	GPT-5.2 (Medium) OpenAI	1,242	52.8%	187.0s	$4.81	15,472
42	GPT-5.1 (High) OpenAI	1,237	55.9%	164.6s	$3.44	8,405
43	MiniMax M2.5 MiniMax	1,235	55.5%	140.5s	$0.44	5,389
44	Grok 4.3 xAI	1,235	49.5%	108.3s	$1.56	29,275
45	GPT-5.2 (None) OpenAI	1,235	52.0%	180.6s	$4.81	16,146
46	Claude Opus 4.1 (Thinking) Anthropic	1,230	58.5%	111.6s	$30	7,473
47	Claude Opus 4 Anthropic	1,229	59.9%	91.3s	$30	7,932
48	Grok 4.20 Beta xAI	1,226	50.0%	60.9s	$1.56	17,286
49	Claude Sonnet 4.5 Anthropic	1,225	51.6%	96.4s	$6.6	21,320
50	Gemini 3 Flash Preview Google	1,225	58.3%	35.6s	$1.13	9,357
51	Claude Opus 4.1 Anthropic	1,222	51.4%	110.8s	$30	20,469
52	Claude Sonnet 4.5 (Thinking) Anthropic	1,222	50.9%	96.2s	$6.6	20,899
53	GPT-5.3 Codex OpenAI	1,219	51.2%	65.1s	$4.81	11,788
54	GPT-5.1 (Medium) OpenAI	1,217	54.2%	128.4s	$3.44	10,709
55	GPT-5 (Minimal) OpenAI	1,213	52.3%	137.0s	$3.44	17,332
56	GPT-5.1 (Low) OpenAI	1,209	53.3%	102.5s	$3.44	10,832
57	GLM 4.6 Zhipu	1,208	54.6%	109.1s	$0.96	14,578
58	Claude 3.7 Sonnet Anthropic	1,203	54.3%	131.9s	—	7,529
59	DeepSeek-V3.2-Exp DeepSeek	1,203	53.1%	228.0s	$0.31	9,583
60	GLM 4.5 Zhipu	1,203	54.4%	139.7s	$1	9,161
61	Step 3.7 Flash Stepfun	1,200	41.2%	143.1s	$0.44	8,165
62	Qwen3.5 397B A17B Alibaba	1,198	50.1%	142.5s	$1.09	4,149
63	Claude Sonnet 4 (Thinking) Anthropic	1,197	54.9%	74.2s	$6	7,680
64	GPT-5.1 Codex OpenAI	1,197	52.3%	108.9s	$3.44	1,931
65	MiMo-V2-Flash Xiaomi	1,197	48.6%	66.6s	—	16,624
66	GLM 4.7 Flash Zhipu	1,193	49.7%	153.7s	$0.15	5,798
67	MiniMax M2.1 MiniMax	1,193	50.4%	131.5s	$0.53	9,381
68	Claude Sonnet 4 Anthropic	1,192	55.4%	79.0s	$6	8,318
69	DeepSeek-V3.2 DeepSeek	1,191	46.7%	182.6s	$1	14,620
70	GPT-5.1 (None) OpenAI	1,190	50.5%	106.3s	$3.44	10,899
71	Hy3 Low Tencent	1,190	38.5%	88.4s	$0.25	1,187
72	DeepSeek-V3.1 (Thinking) DeepSeek	1,189	52.5%	235.2s	$0.44	7,784
73	GPT-5 mini (Default) OpenAI	1,187	46.4%	138.5s	$0.69	19,371
74	Nemotron 3 Ultra NVIDIA	1,186	39.0%	136.9s	$0	4,868
75	MiniMax M2 Stable MiniMax	1,176	48.1%	131.1s	$0.45	5,275
76	Gemini 2.5 Pro Google	1,170	55.1%	86.9s	$3.44	3,845
77	Qwen3.5 Plus 02-15 Alibaba	1,164	42.7%	136.5s	$0.59	9,623
78	Qwen3 Coder 480B A35B Instruct Alibaba	1,160	59.0%	59.0s	$0.67	902
79	DeepSeek-R1-0528 DeepSeek	1,157	49.5%	107.1s	$1.15	8,357
80	Claude Haiku 4.5 Anthropic	1,155	44.6%	47.2s	$2.2	18,482
81	GLM 4.5 Air Zhipu	1,153	48.4%	117.4s	$0.43	8,103
82	Qwen3 Max Alibaba	1,153	43.9%	121.5s	$1.56	16,354
83	GPT-5.1 Codex Mini OpenAI	1,151	43.5%	75.1s	$0.69	17,161
84	DeepSeek-V3.1 DeepSeek	1,143	47.2%	48.7s	$0.46	9,535
85	Prime Intellect. INTELLECT-3 Prime Intellect	1,142	41.1%	87.2s	—	15,077
86	GPT-4.1 OpenAI	1,138	59.1%	97.2s	$3.5	767
87	Mistral Large 3 (2512) Mistral	1,138	41.5%	113.7s	$0.75	14,978
88	Trinity Large Thinking Arcee Ai	1,134	38.1%	94.7s	$0.39	6,482
89	Grok 4.1 Fast xAI	1,133	40.5%	34.8s	—	16,318
90	Mistral Medium 3.1 (2508) Mistral	1,132	40.7%	121.2s	$0.8	14,173
91	Grok 4 Fast xAI	1,131	40.5%	41.0s	—	18,019
92	GPT-4.1 mini OpenAI	1,129	58.5%	86.6s	$0.7	681
93	Grok 4.1 Fast (Reasoning) xAI	1,127	39.5%	97.5s	—	16,053
94	Gemini 2.5 Flash Preview 09-2025 Google	1,124	44.3%	45.2s	$0.85	9,097
95	DeepSeek-V3-0324 DeepSeek	1,112	43.9%	70.9s	$0.63	9,090
96	Gemini 2.5 Flash Lite Preview 09-2025 Google	1,105	45.9%	14.1s	—	3,139
97	GPT-5 nano (Default) OpenAI	1,104	46.6%	98.0s	$0.14	3,043
98	Ministral 3 14B (2512) Mistral	1,097	43.6%	90.6s	$0.35	1,017
99	o3 OpenAI	1,093	56.9%	53.7s	$3.5	590
100	Grok 3 xAI	1,091	38.8%	51.5s	—	13,192
101	Gemini 3.1 Flash-Lite Preview Google	1,090	34.0%	5.4s	$0.56	12,910
102	Grok 4 Fast (Reasoning) xAI	1,087	35.1%	30.1s	—	18,584
103	Mistral Medium 3 (2505) Mistral	1,077	45.3%	67.1s	$0.8	2,890
104	Grok 4 xAI	1,068	35.6%	82.8s	—	11,878
105	Gemini 2.5 Flash Google	1,062	43.5%	78.2s	$0.85	3,254
106	o4-mini OpenAI	1,061	50.0%	35.8s	$1.93	877
107	Grok Code Fast 1 xAI	1,055	40.1%	19.6s	—	8,460
108	GPT OSS 120B OpenAI	1,052	40.6%	23.3s	$0.26	2,487
109	Devstral Medium Mistral	1,050	41.3%	20.9s	—	3,285
110	Ministral 3 8B (2512) Mistral	1,049	38.7%	51.3s	$0.3	1,034
111	Mercury 2 Inception	1,035	21.2%	3.8s	$0.38	3,857
112	Kimi K2 Moonshot	1,032	46.4%	70.8s	$1	596
113	Olmo 3.1 32B Think Allenai	1,032	28.5%	151.5s	—	8,236
114	GPT-4.1 nano OpenAI	1,029	49.6%	21.7s	$0.18	858
115	Codestral 2508 Mistral	1,025	36.2%	14.9s	$0.45	3,142
116	Qwen3-235B-A22B-Thinking-2507 Alibaba	1,016	34.3%	122.0s	$0.75	2,908
117	Grok 3 Mini xAI	1,014	39.4%	27.0s	—	3,594
118	Qwen3-235B-A22B-Instruct-2507 Alibaba	1,010	35.2%	64.0s	$0.3	3,206
119	Ministral 3 3B (2512) Mistral	1,007	33.0%	34.8s	$0.15	1,253
120	Qwen3-235B-A22B Alibaba	987	33.1%	95.3s	$0.8	2,399
121	GPT-4o OpenAI	965	42.3%	18.3s	$4.38	854
122	Qwen3 30B-A3B Alibaba	956	33.8%	31.9s	$0.24	1,184
123	Mistral Large 2.1 (2411) Mistral	953	37.9%	35.8s	—	584
124	Mistral Small 3.2 Mistral	947	39.4%	14.8s	$0.14	567
125	Llama 3.1 Nemotron Ultra 253B NVIDIA	936	23.4%	33.0s	—	1,464
126	Llama 4 Maverick Meta Llama	896	33.7%	14.3s	$0.35	790
127	Codestral 2 (2501) Mistral	870	27.6%	6.8s	$0.45	702
128	Devstral Small 1.1 Mistral	846	26.7%	9.8s	—	600
129	Llama 4 Scout Meta Llama	831	27.4%	14.9s	$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.