• New Chat
  • Leaderboard
  • Search
Terms of UsePrivacy Policy
Start Voting
Overview
Agent
Start Voting
Agent

Min
Max

Min
Max

Min
Max

Min
Max

Vision Arena📝OCR

View overall rankings across multimodal AI models capable of reasoning over visual inputs.

Jul 1, 2026
481,634 votes
89 models
Rank by
Rank Spread
1
16
Anthropic
claude-fable-5
Anthropic · Proprietary
1327±15
1,793$10 / $501M
2
15
Anthropic
claude-opus-4-7-thinking
Anthropic · Proprietary
1320±7
12,427$5 / $251M
3
19
Anthropic
claude-opus-4-7
Anthropic · Proprietary
1313±7
12,926$5 / $251M
4
19
Anthropic
claude-opus-4-6
Anthropic · Proprietary
1313±7
15,570$5 / $251M
5
19
Anthropic
claude-opus-4-6-thinking
Anthropic · Proprietary
1313±7
12,285$5 / $251M
6
318
gemini-3-pro
Google · Proprietary
1303±8
8,112$2 / $121M
7
219
Meta
muse-spark
Meta · Proprietary
1302±10
4,069N/AN/A
8
318
gpt-5.4-high
OpenAI · Proprietary
1302±7
12,334$2.50 / $151.1M
9
319
Anthropic
claude-opus-4-8-thinking
Anthropic · Proprietary
1301±9
5,644$5 / $251M
10
619
gemini-3.1-pro-preview
Google · Proprietary
1298±6
23,510$2 / $121M
11
621
gpt-5.5
OpenAI · Proprietary
1297±8
10,968$5 / $301.1M
12
622
gpt-5.5-high
OpenAI · Proprietary
1296±8
10,086$5 / $301.1M
13
622
Anthropic
claude-sonnet-4-6
Anthropic · Proprietary
1291±7
15,864$3 / $151M
14
622
Anthropic
claude-opus-4-8
Anthropic · Proprietary
1290±9
5,887$5 / $251M
15
623
gemini-3.5-flash
Google · Proprietary
1290±11
3,476$1.50 / $91M
16
622
gpt-5.4
OpenAI · Proprietary
1289±7
12,217$2.50 / $151.1M
17
622
gpt-5.2-chat-latest-20260210
OpenAI · Proprietary
1288±7
10,852$1.75 / $14128K
18
1122
gemini-3-flash
Google · Proprietary
1285±5
22,305$0.50 / $31M
19
825
gpt-5.5-instant
OpenAI · Proprietary
1283±10
5,469$5 / $301.1M
20
1226
kimi-k2.6
Moonshot · Modified MIT
1281±8
10,297$0.95 / $4262.1K
21
633
Anthropic
claude-sonnet-5-thinking
Anthropic · Proprietary
1279±15
1,599$2 / $101M
22
1129
qwen3.7-plus
Alibaba · Proprietary
1278±11
3,217$0.32 / $1.281M
23
1836
Bytedance
dola-seed-2.0-pro
Bytedance · Proprietary
1270±8
7,360N/AN/A
24
1933
gemini-3-flash (thinking-minimal)
Google · Proprietary
1270±6
21,191$0.50 / $31M
25
1934
gemma-4-31b
Google · Apache 2.0
1269±7
21,332$0.14 / $0.40262.1K
26
2037
gpt-5.4-mini-high
OpenAI · Proprietary
1267±7
14,171$0.75 / $4.50400K
27
2137
qwen3.5-397b-a17b
Alibaba · Apache 2.0
1266±6
14,729$0.39 / $2.45256K
28
2139
kimi-k2.5-thinking
Moonshot · Modified MIT
1263±6
16,332$0.60 / $3N/A
29
2140
grok-4.20-beta-0309-reasoning
xAI · Proprietary
1261±7
14,966$2 / $62M
30
2242
gpt-5.2-high
OpenAI · Proprietary
1259±7
12,396$1.75 / $14400K
31
2243
grok-4.20-multi-agent-beta-0309
xAI · Proprietary
1258±7
13,575$2 / $62M
32
2243
gpt-5.1-high
OpenAI · Proprietary
1257±9
5,732$1.25 / $10400K
33
2542
gemini-2.5-pro
Google · Proprietary
1257±5
32,596$1.25 / $101M
34
2443
grok-4.3
xAI · Proprietary
1256±8
10,455$1.25 / $2.501M
35
2245
minimax-m3
MiniMax · MiniMax Community License
1256±10
5,451$0.60 / $2.40N/A
36
2543
gemma-4-26b-a4b
Google · Apache 2.0
1255±7
14,161N/AN/A
37
2846
mimo-v2.5
Xiaomi · MIT
1252±7
12,778$0.10 / $0.281M
38
2647
gpt-5.1
OpenAI · Proprietary
1251±9
6,459$1.25 / $10400K
39
2947
chatgpt-4o-latest-20250326
OpenAI · Proprietary
1249±7
11,170$5 / $15128K
40
3047
gemini-3.1-flash-lite-preview
Google · Proprietary
1246±6
17,747$0.25 / $1.501M
41
2854
kimi-k2.5-instant
Moonshot · Modified MIT
1246±12
2,590$0.38 / $2.02262.1K
42
3052
gpt-5-chat
OpenAI · Proprietary
1244±9
10,804$1.25 / $10128K
43
3652
glm-5v-turbo
Z.ai · Proprietary
1242±7
16,315$1.20 / $4202.8K
44
3655
qwen3.5-122b-a10b
Alibaba · Apache 2.0
1240±7
8,786$0.26 / $2.08262.1K
45
3256
gemini-2.5-flash-preview-09-2025
Google · Proprietary
1240±12
2,787$0.30 / $2.501M
46
3855
gpt-5.2
OpenAI · Proprietary
1238±7
13,212$1.75 / $14400K
47
3760
ernie-5.0-preview-1220
Baidu · Proprietary
1232±14
1,904N/AN/A
48
4160
mimo-v2-omni
Xiaomi · Proprietary
1231±8
7,300$0.40 / $2262.1K
49
4160
qwen3.5-27b
Alibaba · Apache 2.0
1231±7
13,234$0.20 / $1.56262.1K
50
4160
gpt-5-high
OpenAI · Proprietary
1230±9
10,859$1.25 / $10400K
51
4160
qwen3-vl-235b-a22b-instruct
Alibaba · Apache 2.0
1230±8
7,592$0.20 / $0.88262.1K
52
4362
gpt-4.1-2025-04-14
OpenAI · Proprietary
1226±9
11,589$2 / $81M
53
4760
gemini-2.5-flash
Google · Proprietary
1222±5
26,778$0.30 / $2.501M
54
4663
o3-2025-04-16
OpenAI · Proprietary
1222±8
14,665$2 / $8200K
55
4764
gpt-5.4-nano-high
OpenAI · Proprietary
1217±7
14,252$0.20 / $1.25400K
56
4468
qwen-vl-max-2025-08-13
Alibaba · Proprietary
1216±18
1,148$0.52 / $2.08131.1K
57
4766
mistral-medium-3.5
Mistral · Modified MIT
1215±11
3,697$1.50 / $7.50262.1K
58
4170
Anthropic
claude-sonnet-4-20250514-thinking-32k
Anthropic · Proprietary
1215±21
741$3 / $151M
59
4370
Anthropic
claude-opus-4-20250514-thinking-16k
Anthropic · Proprietary
1215±20
848$15 / $75200K
60
4771
Anthropic
claude-3-7-sonnet-20250219-thinking-32k
Anthropic · Proprietary
1209±19
883$3 / $15200K
61
5368
o4-mini-2025-04-16
OpenAI · Proprietary
1208±9
12,038$1.10 / $4.40200K
62
5470
gpt-4.1-mini-2025-04-14
OpenAI · Proprietary
1206±9
10,677$0.40 / $1.601M
63
5372
qwen3-vl-235b-a22b-thinking
Alibaba · Apache 2.0
1201±16
1,357$0.26 / $2.60131.1K
64
5672
gpt-5-mini-high
OpenAI · Proprietary
1196±10
8,176$0.25 / $2400K
65
5573
Anthropic
claude-opus-4-20250514
Anthropic · Proprietary
1196±16
1,359$15 / $75200K
66
5772
grok-4-1-fast-reasoning
xAI · Proprietary
1193±8
8,837$0.20 / $0.502M
67
5680
Anthropic
claude-sonnet-4-20250514
Anthropic · Proprietary
1188±17
1,135$3 / $151M
68
5974
gemini-2.5-flash-lite-preview-06-17-thinking
Google · Proprietary
1188±9
10,701$0.10 / $0.401M
69
5977
gemini-2.5-flash-lite-preview-09-2025-no-thinking
Google · Proprietary
1186±12
2,799$0.10 / $0.401M
70
5783
Anthropic
claude-3-7-sonnet-20250219
Anthropic · Proprietary
1183±19
910$3 / $15200K
71
6682
grok-4-0709
xAI · Proprietary
1175±9
9,989$3 / $15256K
72
6284
Anthropic
claude-3-5-sonnet-20241022
Anthropic · Proprietary
1174±19
961$3 / $15200K
73
6783
mistral-medium-2508
Mistral · Proprietary
1172±7
13,440$0.40 / $2131.1K
74
6384
glm-4.6v
Z.ai · MIT
1170±17
1,431$0.30 / $0.90131.1K
75
6884
gemini-2.0-flash-001
Google · Proprietary
1165±11
3,750$0.10 / $0.401M
76
6885
Tencent
hunyuan-vision-1.5-thinking
Tencent · Proprietary
1162±16
1,401N/AN/A
77
6985
gemma-3-27b-it
Google · Gemma
1162±10
6,329$0.08 / $0.16131.1K
78
7085
mistral-medium-2505
Mistral · Proprietary
1158±10
4,657$0.40 / $2131.1K
79
6888
glm-4.5v
Z.ai · MIT
1158±17
1,217$0.60 / $1.8065.5K
80
6987
gpt-5-nano-high
OpenAI · Proprietary
1156±15
1,708$0.05 / $0.40400K
81
6988
Stepfun
step-1o-turbo-202506
StepFun · Proprietary
1156±17
1,272N/AN/A
82
7388
Meta
llama-4-maverick-17b-128e-instruct
Meta · Llama 4
1152±12
3,160$0.63 / $1.80131.1K
83
7088
Stepfun
step-3
StepFun · Apache 2.0
1152±17
1,246$0.57 / $1.4265.5K
84
7188
Tencent
hunyuan-large-vision
Tencent · Proprietary
1146±20
811N/AN/A
85
7688
mistral-small-2506
Mistral · Apache 2.0
1142±11
4,171$0.10 / $0.3032K
86
7988
mistral-small-3.1-24b-instruct-2503
Mistral · Apache 2.0
1132±10
7,362$0.10 / $0.3032K
87
8088
Meta
llama-4-scout-17b-16e-instruct
Meta · Llama
1129±12
2,858$0.40 / $0.708.2K
88
7989
Anthropic
claude-3-5-haiku-20241022
Anthropic · Proprietary
1127±20
932$1 / $5200K
89
8889
molmo-2-8b
Ai2 · Apache 2.0
1086±25
789$0.20 / $0.2036.9K

Default Leaderboard Plots

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

Confidence Intervals on Model Strength (via Bootstrapping)

Battle Count for Each Combination of Models (without Ties)

USE CASES

  • Chat with AI
  • Build Apps & Websites
  • Write & Edit Text
  • Search the Web
  • Generate Images
  • Generate Videos
  • Chose any model
  • Compare Models Side by Side

LEADERBOARD RANKINGS

  • Overall
  • Agent
  • Text
  • WebDev
  • Image-to-WebDev
  • Text to Image
  • Image Edit
  • Text to Video
  • Image to Video
  • Video Edit
  • Vision
  • Document
  • Search

COMPANY

  • About Us
  • How It Works
  • Blog
  • Careers
  • Changelog
  • Help Center
  • FAQ

LEGAL

  • Terms
  • Privacy
  • Cookies

FOLLOW

  • X
  • LinkedIn
  • YouTube
  • Discord

© Arena Intelligence 2026