Vision Arena

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

Last Updated

Jan 16, 2026

Total Votes

624,976

Total Models

/

Style Control

	Rank Spread
1	1◄─►2	gemini-3-pro	1300	±10	6,370	Google	Proprietary
2	1◄─►3	gemini-3-flash	1280	±12	3,858	Google	Proprietary
3	2◄─►7	gemini-3-flash (thinking-minimal)	1254	±15	1,724	Google	Proprietary
4	3◄─►7	gpt-5.1-high	1249	±11	4,147	OpenAI	Proprietary
5	3◄─►6	gemini-2.5-pro	1248	±7	76,223	Google	Proprietary
6	3◄─►10	gpt-5.1	1241	±11	4,412	OpenAI	Proprietary
7	4◄─►11	chatgpt-4o-latest-20250326	1235	±6	20,118	OpenAI	Proprietary
8	6◄─►17	gemini-2.5-flash-preview-09-2025	1226	±10	5,307	Google	Proprietary
9	6◄─►18	gpt-4.5-preview-2025-02-27	1226	±11	2,925	OpenAI	Proprietary
10	6◄─►18	gpt-5-chat	1223	±8	43,759	OpenAI	Proprietary
11	7◄─►21	ernie-5.0-preview-1220	1217	±12	3,601	Baidu	Proprietary
12	8◄─►18	o3-2025-04-16	1217	±7	49,622	OpenAI	Proprietary
13	8◄─►21	gpt-4.1-2025-04-14	1214	±7	44,903	OpenAI	Proprietary
14	8◄─►21	gemini-2.5-flash	1213	±7	44,278	Google	Proprietary
15	8◄─►23	qwen3-vl-235b-a22b-instruct	1210	±9	7,585	Alibaba	Apache 2.0
16	9◄─►23	gpt-5-high	1208	±8	37,690	OpenAI	Proprietary
17	8◄─►28	claude-sonnet-4-20250514-thinking-32k	1207	±15	1,379	Anthropic	Proprietary
18	8◄─►28	claude-opus-4-20250514-thinking-16k	1206	±15	1,506	Anthropic	Proprietary
19	12◄─►27	gpt-4.1-mini-2025-04-14	1201	±8	44,117	OpenAI	Proprietary
20	12◄─►28	o4-mini-2025-04-16	1200	±7	44,664	OpenAI	Proprietary
21	12◄─►33	claude-3-7-sonnet-20250219-thinking-32k	1195	±15	1,688	Anthropic	Proprietary
22	15◄─►33	o1-2024-12-17	1193	±10	3,694	OpenAI	Proprietary
23	15◄─►33	claude-opus-4-20250514	1191	±12	2,595	Anthropic	Proprietary
24	17◄─►33	gemini-2.5-flash-lite-preview-06-17-thinking	1188	±8	39,495	Google	Proprietary
25	17◄─►34	qwen3-vl-235b-a22b-thinking	1188	±12	2,670	Alibaba	Apache 2.0
26	17◄─►34	hunyuan-vision-1.5-thinking	1188	±12	2,875	Tencent	Proprietary
27	17◄─►34	claude-sonnet-4-20250514	1187	±13	2,088	Anthropic	Proprietary
28	21◄─►34	grok-4-0709	1183	±8	35,147	xAI	Proprietary
29	21◄─►34	gpt-5-mini-high	1182	±9	31,728	OpenAI	Proprietary
30	18◄─►36	qwen-vl-max-2025-08-13	1181	±12	3,486	Alibaba	Proprietary
31	21◄─►35	gemini-1.5-pro-002	1178	±8	8,902	Google	Proprietary
32	21◄─►37	claude-3-7-sonnet-20250219	1178	±9	4,683	Anthropic	Proprietary
33	21◄─►42	gemini-2.5-flash-lite-preview-09-2025-no-thinking	1173	±10	5,347	Google	Proprietary
34	25◄─►42	gemini-2.0-flash-001	1169	±7	9,928	Google	Proprietary
35	31◄─►47	gpt-4o-2024-05-13	1162	±8	23,273	OpenAI	Proprietary
36	32◄─►47	claude-3-5-sonnet-20241022	1162	±7	10,581	Anthropic	Proprietary
37	33◄─►49	gemma-3-27b-it	1157	±8	18,734	Google	Gemma
38	33◄─►49	mistral-medium-2505	1155	±8	11,607	Mistral	Proprietary
39	33◄─►51	glm-4.5v	1155	±12	3,604	Z.ai	MIT
40	30◄─►51	glm-4.6v	1154	±16	1,837	Z.ai	MIT
41	33◄─►51	step-1o-turbo-202506	1152	±14	2,057	StepFun	Proprietary
42	33◄─►52	hunyuan-large-vision	1152	±16	1,455	Tencent	Proprietary
43	35◄─►51	mistral-medium-2508	1148	±8	39,476	Mistral	Proprietary
44	35◄─►51	claude-3-5-sonnet-20240620	1147	±9	21,624	Anthropic	Proprietary
45	35◄─►51	llama-4-maverick-17b-128e-instruct	1146	±9	7,463	Meta	Llama 4
46	35◄─►54	step-3	1145	±12	3,587	StepFun	Apache 2.0
47	35◄─►54	gpt-5-nano-high	1144	±12	4,353	OpenAI	Proprietary
48	37◄─►57	mistral-small-2506	1139	±9	11,809	Mistral	Apache 2.0
49	37◄─►57	gemini-1.5-flash-002	1139	±9	7,241	Google	Proprietary
50	39◄─►59	gemini-2.0-flash-lite-preview-02-05	1133	±10	3,991	Google	Proprietary
51	39◄─►59	claude-3-5-haiku-20241022	1131	±15	1,592	Anthropic	Proprietary
52	46◄─►59	mistral-small-3.1-24b-instruct-2503	1126	±9	31,315	Mistral	Apache 2.0
53	46◄─►59	llama-4-scout-17b-16e-instruct	1126	±10	6,874	Meta	Llama
54	45◄─►59	step-1o-vision-32k-highres	1124	±12	2,833	StepFun	Proprietary
55	48◄─►59	qwen2.5-vl-72b-instruct	1122	±10	3,768	Alibaba	Qwen
56	48◄─►59	gpt-4o-2024-08-06	1119	±12	3,376	OpenAI	Proprietary
57	50◄─►60	gemini-1.5-pro-001	1118	±11	16,734	Google	Proprietary
58	48◄─►62	qwen2.5-vl-32b-instruct	1116	±15	1,490	Alibaba	Apache 2.0
59	50◄─►62	gpt-4-turbo-2024-04-09	1113	±11	13,391	OpenAI	Proprietary
60	58◄─►64	gpt-4o-mini-2024-07-18	1097	±7	17,347	OpenAI	Proprietary
61	58◄─►64	pixtral-large-2411	1094	±9	5,423	Mistral	MRL
62	57◄─►68	gpt-4.1-nano-2025-04-14	1089	±18	1,211	OpenAI	Proprietary
63	60◄─►66	qwen2-vl-72b	1086	±9	5,937	Alibaba	Qwen
64	60◄─►68	qwen-vl-max-1119	1085	±16	1,422	Alibaba	Proprietary
65	62◄─►69	gemini-1.5-flash-8b-001	1071	±10	6,243	Google	Proprietary
66	63◄─►70	claude-3-opus-20240229	1064	±10	15,565	Anthropic	Proprietary
67	62◄─►70	step-1v-32k	1063	±16	1,534	StepFun	Proprietary
68	63◄─►70	gemini-1.5-flash-001	1060	±11	13,260	Google	Proprietary
69	66◄─►74	molmo-72b-0924	1047	±13	3,048	AI2	Apache 2.0
70	65◄─►77	hunyuan-standard-vision-2024-12-31	1043	±21	809	Tencent	Proprietary
71	69◄─►77	llama-3.2-vision-90b-instruct	1033	±8	8,682	Meta	Llama 3.2
72	69◄─►78	qwen2-vl-7b-instruct	1032	±10	5,766	Aliaba	Apache 2.0
73	70◄─►79	pixtral-12b-2409	1025	±9	7,511	Mistral	Apache 2.0
74	69◄─►80	internvl2-26b	1025	±12	5,148	OpenGVLab	MIT
75	69◄─►82	amazon-nova-lite-v1.0	1021	±15	1,854	Amazon	Proprietary
76	70◄─►82	amazon-nova-pro-v1.0	1020	±13	2,335	Amazon	Proprietary
77	70◄─►81	claude-3-sonnet-20240229	1019	±11	12,314	Anthropic	Proprietary
78	73◄─►84	yi-vision	1004	±18	1,219	01 AI	Proprietary
79	74◄─►84	claude-3-haiku-20240307	1002	±12	13,380	Anthropic	Proprietary
80	72◄─►86	c4ai-aya-vision-32b	1000	±22	847	Cohere	CC-BY-NC-4.0
81	75◄─►84	molmo-7b-d-0924	996	±13	2,815	AI2	Apache 2.0
82	78◄─►84	llama-3.2-vision-11b-instruct	992	±11	4,817	Meta	Llama 3.2
83	76◄─►88	nvila-internal-15b-v1	988	±20	1,077	Nvidia	-
84	78◄─►88	llava-onevision-qwen2-72b-ov	981	±18	1,321	LLaVA	Apache 2.0
85	83◄─►88	llava-v1.6-34b	966	±12	4,531	LLaVA	Apache 2.0
86	82◄─►88	minicpm-v-2_6	965	±15	1,987	OpenBMB	Apache 2.0
87	82◄─►88	cogvlm2-llama3-chat-19b	964	±15	1,991	Zhipu AI	CogVLM2
88	83◄─►88	internvl2-4b	958	±12	3,703	OpenGVLab	MIT
89	89◄─►89	phi-3.5-vision-instruct	922	±15	2,592	Microsoft	MIT
90	90◄─►90	phi-3-vision-128k-instruct	884	±18	1,401	Microsoft	MIT

Vision Arena

Default Leaderboard Plots

Confidence Intervals on Model Strength (via Bootstrapping)

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

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

Battle Count for Each Combination of Models (without Ties)