Vision Arena | Overall

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

Feb 18, 2026

667,186 votes

97 models

	Rank Spread
1	13	gemini-3-pro Google · Proprietary	1288±8	11,913
2	15	gemini-3.1-pro-preview Google · Proprietary	1278±12	3,490
3	15	gemini-3-flash Google · Proprietary	1273±9	10,748
4	29	gemini-3-flash (thinking-minimal) Google · Proprietary	1263±10	8,800
5	211	dola-seed-2.0-preview Bytedance · Proprietary	1263±16	1,571
6	412	gpt-5.1-high OpenAI · Proprietary	1251±9	8,523
7	414	kimi-k2.5-thinking Moonshot · Modified MIT	1250±11	4,540
8	415	gpt-5.2-high OpenAI · Proprietary	1249±11	4,366
9	412	gemini-2.5-pro Google · Proprietary	1248±6	80,382
10	517	gpt-5.1 OpenAI · Proprietary	1241±9	9,336
11	517	kimi-k2.5-instant Moonshot · Modified MIT	1240±12	3,423
12	615	chatgpt-4o-latest-20250326 OpenAI · Proprietary	1239±6	24,034
13	822	gpt-5.2 OpenAI · Proprietary	1230±11	4,739
14	923	gemini-2.5-flash-preview-09-2025 Google · Proprietary	1228±10	4,753
15	824	gpt-4.5-preview-2025-02-27 OpenAI · Proprietary	1228±11	2,925
16	1123	gpt-5-chat OpenAI · Proprietary	1225±8	42,550
17	1125	ernie-5.0-preview-1220 Baidu · Proprietary	1221±12	3,324
18	1325	qwen3-vl-235b-a22b-instruct Alibaba · Apache 2.0	1218±8	11,543
19	1324	o3-2025-04-16 OpenAI · Proprietary	1218±7	48,610
20	1327	gpt-4.1-2025-04-14 OpenAI · Proprietary	1215±7	44,026
21	1427	gemini-2.5-flash Google · Proprietary	1212±6	48,786
22	1629	gpt-5-high OpenAI · Proprietary	1209±8	36,691
23	1334	claude-sonnet-4-20250514-thinking-32k Anthropic · Proprietary	1207±16	1,337
24	1334	claude-opus-4-20250514-thinking-16k Anthropic · Proprietary	1207±15	1,479
25	1833	gpt-4.1-mini-2025-04-14 OpenAI · Proprietary	1203±8	43,400
26	2033	o4-mini-2025-04-16 OpenAI · Proprietary	1202±7	44,221
27	2039	claude-3-7-sonnet-20250219-thinking-32k Anthropic · Proprietary	1195±15	1,649
28	2238	o1-2024-12-17 OpenAI · Proprietary	1195±10	3,694
29	2239	claude-opus-4-20250514 Anthropic · Proprietary	1193±12	2,548
30	2341	qwen3-vl-235b-a22b-thinking Alibaba · Apache 2.0	1188±13	2,419
31	2339	gemini-2.5-flash-lite-preview-06-17-thinking Google · Proprietary	1188±8	38,644
32	2341	claude-sonnet-4-20250514 Anthropic · Proprietary	1188±13	2,044
33	2341	qwen-vl-max-2025-08-13 Alibaba · Proprietary	1185±12	3,374
34	2541	gpt-5-mini-high OpenAI · Proprietary	1183±9	30,881
35	2741	grok-4-0709 xAI · Proprietary	1182±8	34,084
36	2741	gemini-1.5-pro-002 Google · Proprietary	1180±8	8,902
37	2743	claude-3-7-sonnet-20250219 Anthropic · Proprietary	1179±9	4,666
38	2749	hunyuan-vision-1.5-thinking Tencent · Proprietary	1176±12	2,488
39	2849	gemini-2.5-flash-lite-preview-09-2025-no-thinking Google · Proprietary	1173±10	4,792
40	3148	gemini-2.0-flash-001 Google · Proprietary	1172±7	10,500
41	3753	gpt-4o-2024-05-13 OpenAI · Proprietary	1164±8	23,273
42	3753	claude-3-5-sonnet-20241022 Anthropic · Proprietary	1163±7	10,548
43	3155	glm-4.6v Z.ai · MIT	1163±14	2,425
44	3855	gemma-3-27b-it Google · Gemma	1158±8	18,104
45	3855	mistral-medium-2505 Mistral · Proprietary	1158±8	11,416
46	3955	mistral-medium-2508 Mistral · Proprietary	1157±7	42,706
47	3857	glm-4.5v Z.ai · MIT	1156±12	3,493
48	3857	step-1o-turbo-202506 StepFun · Proprietary	1155±14	2,027
49	3860	hunyuan-large-vision Tencent · Proprietary	1152±16	1,423
50	4157	claude-3-5-sonnet-20240620 Anthropic · Proprietary	1148±9	21,624
51	4159	llama-4-maverick-17b-128e-instruct Meta · Llama 4	1147±9	7,297
52	4160	step-3 StepFun · Apache 2.0	1147±12	3,448
53	4162	gpt-5-nano-high OpenAI · Proprietary	1146±12	4,203
54	4364	mistral-small-2506 Mistral · Apache 2.0	1141±9	11,562
55	4364	gemini-1.5-flash-002 Google · Proprietary	1141±9	7,241
56	4765	gemini-2.0-flash-lite-preview-02-05 Google · Proprietary	1136±10	3,991
57	4766	claude-3-5-haiku-20241022 Anthropic · Proprietary	1132±15	1,565
58	5066	llama-4-scout-17b-16e-instruct Meta · Llama	1128±10	6,724
59	5166	mistral-small-3.1-24b-instruct-2503 Mistral · Apache 2.0	1128±9	30,780
60	5066	step-1o-vision-32k-highres StepFun · Proprietary	1126±12	2,833
61	5466	qwen2.5-vl-72b-instruct Alibaba · Qwen	1123±10	3,768
62	5467	gpt-4o-2024-08-06 OpenAI · Proprietary	1120±12	3,376
63	5666	gemini-1.5-pro-001 Google · Proprietary	1120±11	16,734
64	5369	qwen2.5-vl-32b-instruct Alibaba · Apache 2.0	1119±15	1,490
65	5769	gpt-4-turbo-2024-04-09 OpenAI · Proprietary	1114±11	13,391
66	5371	molmo-2-8b AI2 · Apache 2.0	1114±20	1,202
67	6471	gpt-4o-mini-2024-07-18 OpenAI · Proprietary	1099±8	17,347
68	6471	pixtral-large-2411 Mistral · MRL	1096±9	5,423
69	6375	gpt-4.1-nano-2025-04-14 OpenAI · Proprietary	1090±18	1,211
70	6673	qwen2-vl-72b Alibaba · Qwen	1087±10	5,937
71	6675	qwen-vl-max-1119 Alibaba · Proprietary	1087±16	1,422
72	6976	gemini-1.5-flash-8b-001 Google · Proprietary	1072±10	6,243
73	7077	claude-3-opus-20240229 Anthropic · Proprietary	1065±10	15,565
74	6977	step-1v-32k StepFun · Proprietary	1065±16	1,534
75	7077	gemini-1.5-flash-001 Google · Proprietary	1062±11	13,260
76	7381	molmo-72b-0924 AI2 · Apache 2.0	1049±13	3,048
77	7285	hunyuan-standard-vision-2024-12-31 Tencent · Proprietary	1045±21	809
78	7684	llama-3.2-vision-90b-instruct Meta · Llama 3.2	1034±9	8,682
79	7685	qwen2-vl-7b-instruct Aliaba · Apache 2.0	1033±10	5,766
80	7786	pixtral-12b-2409 Mistral · Apache 2.0	1027±9	7,511
81	7687	internvl2-26b OpenGVLab · MIT	1026±12	5,148
82	7689	amazon-nova-lite-v1.0 Amazon · Proprietary	1022±15	1,854
83	7789	amazon-nova-pro-v1.0 Amazon · Proprietary	1021±13	2,335
84	7788	claude-3-sonnet-20240229 Anthropic · Proprietary	1020±11	12,314
85	8091	yi-vision 01 AI · Proprietary	1005±18	1,219
86	8191	claude-3-haiku-20240307 Anthropic · Proprietary	1004±12	13,380
87	7893	c4ai-aya-vision-32b Cohere · CC-BY-NC-4.0	1002±22	847
88	8291	molmo-7b-d-0924 AI2 · Apache 2.0	997±13	2,815
89	8591	llama-3.2-vision-11b-instruct Meta · Llama 3.2	994±11	4,817
90	8395	nvila-internal-15b-v1 Nvidia · -	989±20	1,077
91	8595	llava-onevision-qwen2-72b-ov LLaVA · Apache 2.0	982±18	1,321
92	9095	llava-v1.6-34b LLaVA · Apache 2.0	968±12	4,531
93	8995	minicpm-v-2_6 OpenBMB · Apache 2.0	966±16	1,987
94	8995	cogvlm2-llama3-chat-19b Zhipu AI · CogVLM2	966±15	1,991
95	9095	internvl2-4b OpenGVLab · MIT	960±12	3,703
96	9696	phi-3.5-vision-instruct Microsoft · MIT	923±16	2,592
97	9797	phi-3-vision-128k-instruct Microsoft · MIT	885±18	1,401

Vision Arena | Overall

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)