Skip to content

Benchmark

Arena ELO Vision - English

People compare model answers and choose which they prefer. Higher scores are better.

Models measured
147
Scores recorded
168
Sources
1
IQ weight
1.75

Score distribution

1,193
Mean
1,230.5
Median
107.3
Std dev
1,012.8
10th percentile
1,297.8
90th percentile
836.8
Min
836.81,325.5
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.75 for chat models, measured over 147 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
101Hunyuan Large Vision· LMArena data1,198
102GPT-4.5 Preview2025-02-27· LMArena data1,190.4
103GPT-4.1 mini2025-04-14· LMArena data1,190.3
104Gemini 2.5 Flash-Lite PreviewThinking, 06-17· LMArena data1,189.6
105Hunyuan Vision 1.5 Thinking· LMArena data1,189.5
106Gemma 3 27B IT· LMArena data1,187.7
107Claude Sonnet 4Thinking 32K, 20250514· LMArena data1,186
108Claude Opus 420250514· LMArena data1,185.2
109GLM 4.5V· LMArena data1,183.1
110Claude Opus 4Thinking 16K, 20250514· LMArena data1,182
111Mistral Medium 3.12508· LMArena data1,179
112GLM 4.6v· LMArena data1,178.4
113Gemini 2.0 Flash001· LMArena data1,174.5
114Mistral Medium 32505· LMArena data1,171.5
115GPT-5 nanoHigh· LMArena data1,169.5
116Claude Sonnet 3.7Thinking 32K, 20250219· LMArena data1,163.7
117o12024-12-17· LMArena data1,162
118Mistral Small 3.22506· LMArena data1,161.3
119Gemini 1.5 Pro002· LMArena data1,158.8
120Claude Sonnet 420250514· LMArena data1,158.4
121Qwen 2.5 VL 32B Instruct· LMArena data1,158.2
122Gemini 1.5 Flash002· LMArena data1,156.9
123Mistral Small 3.1 24B Instruct2503· LMArena data1,156.6
124Llama 4 Maverick· LMArena data1,149.8
125GPT 4o2024-05-13· LMArena data1,146.8
126Step-1o Vision 32KHigh Res· LMArena data1,146.6
127Llama 4 Scout· LMArena data1,143
128Claude Sonnet 3.720250219· LMArena data1,140.5
129Claude 3.5 Sonnet20241022· LMArena data1,123.5
130Gemini 2.0 Flash-Lite Preview02-05· LMArena data1,120.8
131Claude 3.5 Sonnet20240620· LMArena data1,118.6
132Qwen2.5 VL 72B Instruct· LMArena data1,114.6
133GPT-4.1 nano2025-04-14· LMArena data1,100.5
134Pixtral Large2411· LMArena data1,099.8
135GPT-4 Turbo2024-04-09· LMArena data1,096.3
136Molmo 2 8B· LMArena data1,095.3
137Gemini 1.5 Pro001· LMArena data1,089.2
138Claude Haiku 3.520241022· LMArena data1,086.6
139GPT-4o mini2024-07-18· LMArena data1,072.2
140Qwen-VL Max1119· LMArena data1,071.8
141GPT 4o2024-08-06· LMArena data1,071.7
142Step-1V 32K· LMArena data1,071.3
143Gemini 1.5 Flash-8B001· LMArena data1,070.1
144Qwen2-VL-72B· LMArena data1,053.5
145Molmo 72B0924· LMArena data1,044.9
146InternVL2 26B· LMArena data1,037
147Gemini 1.5 Flash001· LMArena data1,026.8
148Pixtral 12B2409· LMArena data1,026.4
149Hunyuan Standard Vision2024-12-31· LMArena data1,019.5
150Claude 3 Opus20240229· LMArena data1,018.2

Showing 101–150 of 168 results.

Data credits: LMArena, CC BY 4.0 (licence)