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
51Gemini 3.5 Flash Lite· LMArena data1,268.1
52Grok 4.6High· LMArena data1,267.9
53Grok 4.20Reasoning, 0309, beta· LMArena data1,266.3
54Gemini 2.5 Flash Preview09-2025· LMArena data1,263.1
55Grok 4.20 Multi-Agent0309, beta· LMArena data1,262.9
56Gemini 2.5 Pro· LMArena data1,260.5
57GPT-5.6 LunaxHigh· LMArena data1,259.9
58MiniMax M3· LMArena data1,259.8
59MiMo V2.5· LMArena data1,259.7
60Kimi K2.5Instant· LMArena data1,259
61Gemini 3 FlashMinimal· LMArena data1,258.2
62Mistral Large 4· LMArena data1,257.7
63GPT-6.1 TerraMax· LMArena data1,255.3
64Gemma 4 26B A4B· LMArena data1,255.3
65Qwen3 VL 235B· LMArena data1,255.1
66GPT-5.5 Instant· LMArena data1,254.6
67GPT 5.1High· LMArena data1,250.8
68Qwen3.5 122B-A10B· LMArena data1,248.7
69ERNIE-5.0-Preview1220· LMArena data1,247.9
70Qwen3.5 27B· LMArena data1,245.5
71GPT-5.4 miniHigh· LMArena data1,243.9
72Inkling Small· LMArena data1,243.8
73GPT 5.2High· LMArena data1,243.6
74ChatGPT-4o Latest20250326· LMArena data1,243.6
75MiMo V2 Omni· LMArena data1,242.9
76Mistral Medium 3.5· LMArena data1,241.9
77Gemini 2.5 Flash· LMArena data1,239.1
78NVIDIA Nemotron 3.5 Super VL 120BNVFP4· LMArena data1,238.5
79Grok 4.3· LMArena data1,235.9
80Mistral Large 3· LMArena data1,233.8
81Gemini 3.1 Flash Lite· LMArena data1,232.7
82GPT-5 Chat· LMArena data1,231.5
83GPT 5.2· LMArena data1,230.6
84GPT 5.1· LMArena data1,230.5
85Grok 4.7xHigh· LMArena data1,229.6
86GPT-6 SolMax· LMArena data1,226.7
87Qwen-VL Max2025-08-13· LMArena data1,226.4
88Grok 40709· LMArena data1,221.7
89OpenAI o32025-04-16· LMArena data1,220.8
90Grok 4.1 FastReasoning· LMArena data1,219.4
91GPT 5High· LMArena data1,217.8
92GPT-5 miniHigh· LMArena data1,216.2
93Qwen3 VL 235Bthinking· LMArena data1,215.3
94Step-1o Turbo202506· LMArena data1,212.1
95GPT 4.12025-04-14· LMArena data1,208.9
96GPT-6 LunaMax· LMArena data1,207.6
97GPT 5.4 NanoHigh· LMArena data1,201.7
98Gemini 2.5 Flash-Lite PreviewNon-Thinking, 09-2025· LMArena data1,200.5
99o4 Mini2025-04-16· LMArena data1,199.6
100Step3· LMArena data1,198.1

Showing 51–100 of 168 results.

Data credits: LMArena, CC BY 4.0 (licence)