Vision Arena

View rankings across multimodal, generative AI models capable of understanding and processing visual inputs

Last Updated

Jan 29, 2026

Total Votes

641,447

Total Models

/

Style Control

	Rank Spread
1	1◄─►2	gemini-3-pro	1293	±9	9,161	Google	Proprietary
2	1◄─►3	gemini-3-flash	1281	±10	6,460	Google	Proprietary
3	3◄─►8	gemini-3-flash (thinking-minimal)	1258	±11	5,089	Google	Proprietary
4	2◄─►12	gpt-5.2-high	1253	±18	1,394	OpenAI	Proprietary
5	3◄─►10	gpt-5.1-high	1251	±10	5,984	OpenAI	Proprietary
6	3◄─►12	kimi-k2.5-thinking	1249	±17	1,458	Moonshot	Modified MIT
7	3◄─►9	gemini-2.5-pro	1248	±6	77,676	Google	Proprietary
8	3◄─►14	gpt-5.1	1238	±10	6,479	OpenAI	Proprietary
9	5◄─►14	chatgpt-4o-latest-20250326	1235	±6	21,942	OpenAI	Proprietary
10	6◄─►21	gemini-2.5-flash-preview-09-2025	1226	±10	5,312	Google	Proprietary
11	6◄─►21	gpt-4.5-preview-2025-02-27	1226	±11	2,925	OpenAI	Proprietary
12	4◄─►24	gpt-5.2	1225	±19	1,470	OpenAI	Proprietary
13	8◄─►21	gpt-5-chat	1223	±7	43,537	OpenAI	Proprietary
14	8◄─►24	ernie-5.0-preview-1220	1219	±12	3,637	Baidu	Proprietary
15	10◄─►21	o3-2025-04-16	1216	±7	49,431	OpenAI	Proprietary
16	10◄─►24	gpt-4.1-2025-04-14	1213	±7	44,691	OpenAI	Proprietary
17	10◄─►24	gemini-2.5-flash	1213	±7	45,888	Google	Proprietary
18	10◄─►26	qwen3-vl-235b-a22b-instruct	1209	±9	9,436	Alibaba	Apache 2.0
19	10◄─►26	gpt-5-high	1208	±8	37,812	OpenAI	Proprietary
20	10◄─►31	claude-sonnet-4-20250514-thinking-32k	1206	±15	1,373	Anthropic	Proprietary
21	10◄─►31	claude-opus-4-20250514-thinking-16k	1206	±15	1,501	Anthropic	Proprietary
22	14◄─►30	gpt-4.1-mini-2025-04-14	1201	±8	43,914	OpenAI	Proprietary
23	14◄─►31	o4-mini-2025-04-16	1200	±7	44,485	OpenAI	Proprietary
24	14◄─►36	claude-3-7-sonnet-20250219-thinking-32k	1194	±15	1,684	Anthropic	Proprietary
25	18◄─►36	o1-2024-12-17	1193	±10	3,694	OpenAI	Proprietary
26	18◄─►36	claude-opus-4-20250514	1191	±12	2,589	Anthropic	Proprietary
27	20◄─►36	gemini-2.5-flash-lite-preview-06-17-thinking	1188	±8	39,346	Google	Proprietary
28	20◄─►38	hunyuan-vision-1.5-thinking	1187	±12	2,878	Tencent	Proprietary
29	20◄─►38	qwen3-vl-235b-a22b-thinking	1187	±12	2,672	Alibaba	Apache 2.0
30	20◄─►38	claude-sonnet-4-20250514	1186	±13	2,078	Anthropic	Proprietary
31	24◄─►38	grok-4-0709	1182	±8	34,969	xAI	Proprietary
32	24◄─►38	gpt-5-mini-high	1182	±9	31,608	OpenAI	Proprietary
33	21◄─►39	qwen-vl-max-2025-08-13	1181	±12	3,474	Alibaba	Proprietary
34	24◄─►38	gemini-1.5-pro-002	1178	±8	8,902	Google	Proprietary
35	24◄─►40	claude-3-7-sonnet-20250219	1177	±9	4,680	Anthropic	Proprietary
36	24◄─►45	gemini-2.5-flash-lite-preview-09-2025-no-thinking	1173	±10	5,343	Google	Proprietary
37	28◄─►45	gemini-2.0-flash-001	1170	±7	9,905	Google	Proprietary
38	28◄─►50	glm-4.6v	1163	±14	2,616	Z.ai	MIT
39	34◄─►50	gpt-4o-2024-05-13	1162	±8	23,273	OpenAI	Proprietary
40	35◄─►50	claude-3-5-sonnet-20241022	1161	±7	10,575	Anthropic	Proprietary
41	36◄─►52	gemma-3-27b-it	1156	±8	18,648	Google	Gemma
42	36◄─►52	mistral-medium-2505	1155	±8	11,565	Mistral	Proprietary
43	36◄─►54	glm-4.5v	1154	±12	3,595	Z.ai	MIT
44	36◄─►54	step-1o-turbo-202506	1152	±14	2,052	StepFun	Proprietary
45	36◄─►54	hunyuan-large-vision	1152	±16	1,447	Tencent	Proprietary
46	38◄─►54	mistral-medium-2508	1149	±7	40,990	Mistral	Proprietary
47	38◄─►54	claude-3-5-sonnet-20240620	1146	±9	21,624	Anthropic	Proprietary
48	38◄─►54	llama-4-maverick-17b-128e-instruct	1145	±9	7,440	Meta	Llama 4
49	38◄─►57	gpt-5-nano-high	1145	±11	4,347	OpenAI	Proprietary
50	38◄─►57	step-3	1144	±12	3,575	StepFun	Apache 2.0
51	41◄─►60	gemini-1.5-flash-002	1139	±9	7,241	Google	Proprietary
52	41◄─►60	mistral-small-2506	1139	±9	11,764	Mistral	Apache 2.0
53	43◄─►62	gemini-2.0-flash-lite-preview-02-05	1133	±10	3,991	Google	Proprietary
54	43◄─►62	claude-3-5-haiku-20241022	1130	±15	1,590	Anthropic	Proprietary
55	49◄─►62	mistral-small-3.1-24b-instruct-2503	1127	±9	31,139	Mistral	Apache 2.0
56	49◄─►62	llama-4-scout-17b-16e-instruct	1125	±10	6,846	Meta	Llama
57	49◄─►62	step-1o-vision-32k-highres	1123	±12	2,833	StepFun	Proprietary
58	51◄─►62	qwen2.5-vl-72b-instruct	1121	±10	3,768	Alibaba	Qwen
59	51◄─►62	gpt-4o-2024-08-06	1118	±12	3,376	OpenAI	Proprietary
60	53◄─►63	gemini-1.5-pro-001	1118	±11	16,734	Google	Proprietary
61	51◄─►65	qwen2.5-vl-32b-instruct	1116	±15	1,490	Alibaba	Apache 2.0
62	53◄─►65	gpt-4-turbo-2024-04-09	1113	±11	13,391	OpenAI	Proprietary
63	61◄─►67	gpt-4o-mini-2024-07-18	1097	±7	17,347	OpenAI	Proprietary
64	61◄─►67	pixtral-large-2411	1094	±9	5,423	Mistral	MRL
65	60◄─►71	gpt-4.1-nano-2025-04-14	1088	±18	1,211	OpenAI	Proprietary
66	63◄─►69	qwen2-vl-72b	1085	±9	5,937	Alibaba	Qwen
67	63◄─►71	qwen-vl-max-1119	1085	±16	1,422	Alibaba	Proprietary
68	65◄─►72	gemini-1.5-flash-8b-001	1070	±10	6,243	Google	Proprietary
69	66◄─►73	claude-3-opus-20240229	1064	±10	15,565	Anthropic	Proprietary
70	65◄─►73	step-1v-32k	1063	±16	1,534	StepFun	Proprietary
71	66◄─►73	gemini-1.5-flash-001	1060	±11	13,260	Google	Proprietary
72	69◄─►77	molmo-72b-0924	1047	±13	3,048	AI2	Apache 2.0
73	68◄─►80	hunyuan-standard-vision-2024-12-31	1043	±21	809	Tencent	Proprietary
74	72◄─►80	llama-3.2-vision-90b-instruct	1032	±8	8,682	Meta	Llama 3.2
75	72◄─►81	qwen2-vl-7b-instruct	1032	±10	5,766	Aliaba	Apache 2.0
76	73◄─►82	pixtral-12b-2409	1025	±9	7,511	Mistral	Apache 2.0
77	72◄─►83	internvl2-26b	1024	±12	5,148	OpenGVLab	MIT
78	72◄─►85	amazon-nova-lite-v1.0	1020	±15	1,854	Amazon	Proprietary
79	73◄─►85	amazon-nova-pro-v1.0	1020	±13	2,335	Amazon	Proprietary
80	73◄─►84	claude-3-sonnet-20240229	1019	±11	12,314	Anthropic	Proprietary
81	76◄─►87	yi-vision	1004	±18	1,219	01 AI	Proprietary
82	77◄─►87	claude-3-haiku-20240307	1002	±12	13,380	Anthropic	Proprietary
83	75◄─►89	c4ai-aya-vision-32b	1000	±22	847	Cohere	CC-BY-NC-4.0
84	78◄─►87	molmo-7b-d-0924	996	±13	2,815	AI2	Apache 2.0
85	81◄─►87	llama-3.2-vision-11b-instruct	992	±11	4,817	Meta	Llama 3.2
86	79◄─►91	nvila-internal-15b-v1	988	±20	1,077	Nvidia	-
87	81◄─►91	llava-onevision-qwen2-72b-ov	980	±18	1,321	LLaVA	Apache 2.0
88	86◄─►91	llava-v1.6-34b	966	±12	4,531	LLaVA	Apache 2.0
89	85◄─►91	minicpm-v-2_6	965	±15	1,987	OpenBMB	Apache 2.0
90	85◄─►91	cogvlm2-llama3-chat-19b	964	±15	1,991	Zhipu AI	CogVLM2
91	86◄─►91	internvl2-4b	958	±12	3,703	OpenGVLab	MIT
92	92◄─►92	phi-3.5-vision-instruct	922	±15	2,592	Microsoft	MIT
93	93◄─►93	phi-3-vision-128k-instruct	884	±18	1,401	Microsoft	MIT

Vision Arena

Default Leaderboard Plots

Average Win Rate Against All Other Models (Uniform Sampling and No Ties)

Confidence Intervals on Model Strength (via Bootstrapping)

Fraction of Model A Wins for All Non-tied A vs. B Battles

Battle Count for Each Combination of Models (without Ties)