⌕Is ___ SOTA?
benchmarks / lmarena-vision

LMArena Vision (style-controlled)

Arena score. Status display_only. Contamination risk low. Maintainer LMArena. Source.

Display only.This benchmark cannot decide a crown under methodology v1.0.
ModelScoreDateKind
GPT-4O1171.72117844754762025-06-17snapshot
Claude 3.5 Sonnet1162.70942384751562025-05-22snapshot
Gemini 1.5 Pro Api 05141122.19836842026962024-10-15snapshot
GPT-4 Turbo1112.18458004365582025-02-17snapshot
GPT-4O Mini1099.82847537876982025-02-06snapshot
Qwen2 VL 72B1092.93703163295142025-01-15snapshot
Gemini 1.5 Flash Api 05141071.84673338811722024-10-15snapshot
Claude 3 Opus1058.39378452869192024-10-28snapshot
Llama 3.2 Vision 90B Instruct1046.49083326926982024-10-28snapshot
Claude 3 Sonnet1025.4372478712942024-10-28snapshot
Qwen2 VL 7B Instruct1031.2629624083962024-10-28snapshot
Internvl2 26B1021.27188802100072024-10-15snapshot
Gemini 1.5 Pro 0011122.9936007444672025-03-17snapshot
Gemini 1.5 Flash 0011072.37344775685942024-11-22snapshot
Gemini 1.5 Flash 8B 0011075.82940139972152024-12-15snapshot
Gemini 1.5 Pro 0021180.73109377443822025-07-07snapshot
Gemini 1.5 Flash 0021138.33279421094182025-04-23snapshot
Pixtral Large 24111095.4057525032672025-02-03snapshot
Qwen VL Max 11191095.23872888903042025-02-06snapshot
Step 1O Vision 32k Highres1123.59342644833372025-03-25snapshot
Gemini 2.0 Flash 0011166.9402997519212025-05-22snapshot
Gemini 2.0 Flash Lite Preview 02.051135.3200179976222025-04-16snapshot
Qwen2 5 VL 72B Instruct1120.64743589726462025-03-03snapshot
O11188.37419760329892025-08-28snapshot
Claude 3.7 Sonnet1174.32977891858852025-06-18snapshot
GPT-4.5 Preview1225.2139649677932026-02-06snapshot
ChatGPT 4O Latest1239.7808963014542026-03-13snapshot
GPT-4.11214.62371494810492026-01-09snapshot
GPT-4.1 Mini1196.86719999511022025-09-18snapshot
O31216.83713550836752026-01-09snapshot