Benchmark

Vision Arena (arena.ai)

Measures: Users upload an image with a prompt, two anonymous multimodal models answer, and the user votes; votes are turned into an Elo-scale score with a confidence interval, style-controlled by default.

Maintainer
Arena Intelligence (arena.ai; formerly LMArena / LMSYS Chatbot Arena)
Introduced
2024
Status
Active
Saturation
Headroom
Typical provenance
Independent
In scoring set
Yes

PaperOfficial leaderboard

Leaderboard

Independent and third-party results only.

Leaderboard
ModelArena score (Elo scale)Score CI (+/-)VotesProvenanceSourceEvaluated
Claude Fable 5
Anthropic · High effort
1310—13483IndependentArena28 Sep 2026
Qwen3.8-Max
Alibaba Group · Max effort
1301—10086IndependentArena28 Sep 2026
Claude Opus 4.6
Anthropic · High effort
1299—22350IndependentArena28 Sep 2026
Claude Opus 4.71299—22948IndependentArena28 Sep 2026
Claude Opus 4.7
Anthropic · High effort
1298—22559IndependentArena28 Sep 2026
Gemini 3.7 Flash
Google DeepMind · High effort
1295—3425IndependentArena28 Sep 2026
Claude Opus 4.61295—26904IndependentArena28 Sep 2026
Muse Spark 1.3
Meta · Max effort
1290—3472IndependentArena28 Sep 2026
Gemini 3 Pro1289—13624IndependentArena28 Sep 2026
Claude Opus 5
Anthropic · High effort
1289—15304IndependentArena28 Sep 2026
GPT-5.6 Sol
OpenAI · Xhigh effort
1287—10063IndependentArena28 Sep 2026
Claude Fable 5.1
Anthropic · Max effort
1287—4050IndependentArena28 Sep 2026
GPT-5.51287—26870IndependentArena28 Sep 2026
Claude Opus 4.8
Anthropic · High effort
1286—18020IndependentArena28 Sep 2026
Gemini 3.8 Flash
Google DeepMind · High effort
1286—2711IndependentArena28 Sep 2026
GPT-6 Astra
OpenAI · Max effort
1285—2869IndependentArena28 Sep 2026
Gemini 3.5 Flash
Google DeepMind · High effort
1285—11983IndependentArena28 Sep 2026
GPT-5.4
OpenAI · High effort
1284—28881IndependentArena28 Sep 2026
Gemini 3.5 Flash
Google DeepMind · Medium effort
1284—12124IndependentArena28 Sep 2026
GLM-5.3-Flash1281—4832IndependentArena28 Sep 2026
GPT-5.5
OpenAI · High effort
1281—25170IndependentArena28 Sep 2026
Gemini 3.6 Flash
Google DeepMind · High effort
1280—7076IndependentArena28 Sep 2026
Gemini 3.1 Pro1280—45204IndependentArena28 Sep 2026
Muse Spark 1.11280—10906IndependentArena28 Sep 2026
GPT-5.41279—22854IndependentArena28 Sep 2026
Grok 4.51279—10681IndependentArena28 Sep 2026
Claude Opus 4.81279—18583IndependentArena28 Sep 2026
GPT-5.2 Instant (gpt-5.2-chat-latest)1278—16492IndependentArena28 Sep 2026
Claude Sonnet 4.61275—27465IndependentArena28 Sep 2026
Gemini 3 Flash1273—39181IndependentArena28 Sep 2026
GPT-5.6 Terra
OpenAI · Xhigh effort
1268—9990IndependentArena28 Sep 2026
Qwen3.7-Plus1266—12799IndependentArena28 Sep 2026
Kimi K2.61265—16416IndependentArena28 Sep 2026
Claude Sonnet 5
Anthropic · High effort
1265—12722IndependentArena28 Sep 2026
Gemini 3.5 Flash-Lite1264—7200IndependentArena28 Sep 2026
Grok 4.6
SpaceXAI (xAI) · High effort
1263—5374IndependentArena28 Sep 2026
Gemma 4 31B1263—38935IndependentArena28 Sep 2026
GPT-5.6 Luna
OpenAI · Xhigh effort
1259—10099IndependentArena28 Sep 2026
Grok 4.20
SpaceXAI (xAI) · Reasoning effort
1256—26870IndependentArena28 Sep 2026
MiMo-V2.6-Flash1253—1865IndependentArena28 Sep 2026
Kimi K2.5
Moonshot AI · Thinking effort
1252—28367IndependentArena28 Sep 2026
GPT-5.1
OpenAI · High effort
1251—9843IndependentArena28 Sep 2026
GPT-5.4 mini
OpenAI · High effort
1251—25775IndependentArena28 Sep 2026
Grok 4.20 Multi-Agent1250—24597IndependentArena28 Sep 2026
Gemini 2.5 Pro1247—88831IndependentArena28 Sep 2026
MiMo-V2.6-Pro1246—1802IndependentArena28 Sep 2026
Qwen3.5-397B-A17B1245—31178IndependentArena28 Sep 2026
Grok 4.31244—25944IndependentArena28 Sep 2026
GPT-5.2
OpenAI · High effort
1243—19051IndependentArena28 Sep 2026
Qwen3.8-27B1242—6147IndependentArena28 Sep 2026
Gemma 4 26B A4B1242—23977IndependentArena28 Sep 2026
GPT-5.11240—10798IndependentArena28 Sep 2026
ChatGPT-4o1239—23429IndependentArena28 Sep 2026
MiniMax-M31237—17970IndependentArena28 Sep 2026
Gemini 3.1 Flash-Lite1235—27719IndependentArena28 Sep 2026
MiMo-V2.51232—27858IndependentArena28 Sep 2026
Gemini 2.5 Flash1230—4793IndependentArena28 Sep 2026
GLM-5V-Turbo1230—41294IndependentArena28 Sep 2026
GPT-5.21229—20149IndependentArena28 Sep 2026
Qwen3.5-122B-A10B1226—13259IndependentArena28 Sep 2026
GPT-5 Instant (gpt-5-chat-latest)1225—39918IndependentArena28 Sep 2026
GPT-4.51225—2925IndependentArena28 Sep 2026
Qwen3.5-27B1217—28689IndependentArena28 Sep 2026
MiMo-V2-Omni1216—10687IndependentArena28 Sep 2026
o31216—45606IndependentArena28 Sep 2026
Gemini 2.5 Flash1215—60141IndependentArena28 Sep 2026
Qwen3-VL-235B-A22B-Instruct1215—12734IndependentArena28 Sep 2026
GPT-4.11214—41027IndependentArena28 Sep 2026
GPT-5
OpenAI · High effort
1210—34464IndependentArena28 Sep 2026
Inkling Small1207—6658IndependentArena28 Sep 2026
GPT-4.1 mini1201—40235IndependentArena28 Sep 2026
o4-mini1201—41354IndependentArena28 Sep 2026
Mistral Large 31200—7880IndependentArena28 Sep 2026
Mistral Medium 3.51200—5391IndependentArena28 Sep 2026
GPT-5.4 nano
OpenAI · High effort
1200—25685IndependentArena28 Sep 2026
Grok 4.1 Fast
SpaceXAI (xAI) · Reasoning effort
1196—13401IndependentArena28 Sep 2026
o11193—3694IndependentArena28 Sep 2026
Claude Opus 41188—2177IndependentArena28 Sep 2026
Gemini 2.5 Flash-Lite
Google DeepMind · Thinking effort
1188—35953IndependentArena28 Sep 2026
Qwen3-VL-235B-A22B-Thinking
Alibaba Group · Thinking effort
1188—2389IndependentArena28 Sep 2026
Claude Sonnet 41188—1808IndependentArena28 Sep 2026
Qwen-VL-Max1185—3173IndependentArena28 Sep 2026
Grok 41184—30195IndependentArena28 Sep 2026
GPT-5 mini
OpenAI · High effort
1183—28971IndependentArena28 Sep 2026
Gemini 1.5 Pro1179—8902IndependentArena28 Sep 2026
Claude 3.7 Sonnet1175—4519IndependentArena28 Sep 2026
Gemini 2.0 Flash1172—10137IndependentArena28 Sep 2026
GPT-4o1162—23273IndependentArena28 Sep 2026
GLM-4.6V1161—2455IndependentArena28 Sep 2026
Claude 3.5 Sonnet1161—10376IndependentArena28 Sep 2026
Gemma 3 27B1160—16991IndependentArena28 Sep 2026
Mistral Medium 3.11159—41647IndependentArena28 Sep 2026
Mistral Medium 31155—10810IndependentArena28 Sep 2026
GLM-4.5V1153—3249IndependentArena28 Sep 2026
GPT-5 nano
OpenAI · High effort
1146—3934IndependentArena28 Sep 2026
Llama 4 Maverick1145—6925IndependentArena28 Sep 2026
Claude 3.5 Sonnet1145—21624IndependentArena28 Sep 2026
Step31145—3199IndependentArena28 Sep 2026
Gemini 1.5 Flash1141—7241IndependentArena28 Sep 2026
Mistral Small 3.21140—10848IndependentArena28 Sep 2026
Gemini 2.0 Flash-Lite1136—3991IndependentArena28 Sep 2026
Llama 4 Scout1127—6455IndependentArena28 Sep 2026
Mistral Small 3.11127—28507IndependentArena28 Sep 2026
Claude 3.5 Haiku1127—1400IndependentArena28 Sep 2026
Qwen2.5-VL-72B-Instruct1121—3768IndependentArena28 Sep 2026
Gemini 1.5 Pro1120—16734IndependentArena28 Sep 2026
Qwen2.5-VL-32B-Instruct1119—1490IndependentArena28 Sep 2026
GPT-4o1118—3376IndependentArena28 Sep 2026
GPT-4 Turbo1112—13391IndependentArena28 Sep 2026
GPT-4o mini1097—17341IndependentArena28 Sep 2026
Pixtral Large1095—5423IndependentArena28 Sep 2026
GPT-4.1 nano1089—1211IndependentArena28 Sep 2026
Gemini 1.5 Flash-8B1071—6243IndependentArena28 Sep 2026
Step-1V1064—1534IndependentArena28 Sep 2026
Claude 3 Opus1062—15565IndependentArena28 Sep 2026
Gemini 1.5 Flash1060—13260IndependentArena28 Sep 2026
Molmo 72B1045—3048IndependentArena28 Sep 2026
Qwen2-VL-7B-Instruct1031—5766IndependentArena28 Sep 2026
Pixtral 12B1025—7511IndependentArena28 Sep 2026
InternVL2-26B1025—5148IndependentArena28 Sep 2026
Claude 3 Sonnet1016—12314IndependentArena28 Sep 2026
Aya Vision 32B1000—847IndependentArena28 Sep 2026
Claude 3 Haiku1000—13380IndependentArena28 Sep 2026
Molmo 7B-D994—2815IndependentArena28 Sep 2026
CogVLM2-Llama3-Chat-19B964—1991IndependentArena28 Sep 2026
MiniCPM-V-2.6963—1987IndependentArena28 Sep 2026
InternVL2-4B958—3703IndependentArena28 Sep 2026
Phi-3.5-vision920—2592IndependentArena28 Sep 2026
Phi-3-vision881—1401IndependentArena28 Sep 2026
  • Arena (arena.ai), lmarena-ai/leaderboard-dataset, CC BY 4.0.

Sources

Vision Arena (arena.ai): leaderboard and what it measures · supermarket.dev