Skip to content

Benchmark

Arena ELO Vision

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

Models measured
147
Scores recorded
168
Sources
1
IQ weight
1.77

Score distribution

1,185.4
Mean
1,222.2
Median
114.5
Std dev
999
10th percentile
1,304.2
90th percentile
811.6
Min
811.61,325
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.77 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
51Step 5 Preview· LMArena data1,264.5
52Gemini 2.5 Pro· LMArena data1,263.2
53Grok 4.20Reasoning, 0309, beta· LMArena data1,262.6
54Qwen3.5 397B A17B· LMArena data1,262.2
55Grok 4.20 Multi-Agent0309, beta· LMArena data1,259.4
56Gemma 4 26B A4B· LMArena data1,259.2
57MiMo V2.6 Flash· LMArena data1,257.5
58GPT-5.6 LunaxHigh· LMArena data1,256.1
59Gemini 2.5 Flash Preview09-2025· LMArena data1,253.2
60Kimi K2.5Instant· LMArena data1,253
61GPT-5.5 Instant· LMArena data1,250.6
62GPT 5.1High· LMArena data1,249
63MiniMax M3· LMArena data1,248.4
64ERNIE-5.0-Preview1220· LMArena data1,248
65MiMo V2.5· LMArena data1,247.2
66GPT-6.1 TerraMax· LMArena data1,246.9
67Qwen3 VL 235B· LMArena data1,246.6
68Mistral Large 4· LMArena data1,246.1
69GPT 5.2High· LMArena data1,245.7
70GPT-5.4 miniHigh· LMArena data1,244.5
71Qwen3.5 122B-A10B· LMArena data1,244
72ChatGPT-4o Latest20250326· LMArena data1,243.1
73Gemini 3.1 Flash Lite· LMArena data1,240
74Qwen3.5 27B· LMArena data1,239.3
75Inkling Small· LMArena data1,236.3
76Gemini 2.5 Flash· LMArena data1,235.4
77GPT 5.1· LMArena data1,235.1
78GPT-6 SolMax· LMArena data1,232.1
79GPT-5 Chat· LMArena data1,231.9
80Grok 4.7xHigh· LMArena data1,231
81GPT 5.2· LMArena data1,228.3
82Grok 4.3· LMArena data1,228
83MiMo V2 Omni· LMArena data1,227.9
84Mistral Medium 3.5· LMArena data1,222.2
85Mistral Large 3· LMArena data1,222.2
86OpenAI o32025-04-16· LMArena data1,214
87NVIDIA Nemotron 3.5 Super VL 120BNVFP4· LMArena data1,213.4
88Qwen-VL Max2025-08-13· LMArena data1,212
89GPT 4.12025-04-14· LMArena data1,210.5
90Grok 40709· LMArena data1,210.1
91GPT-6 LunaMax· LMArena data1,209.7
92GPT 5High· LMArena data1,208.1
93Qwen3 VL 235Bthinking· LMArena data1,206.7
94GPT-5 miniHigh· LMArena data1,202.6
95Grok 4.1 FastReasoning· LMArena data1,201.5
96Gemini 2.5 Flash-Lite PreviewNon-Thinking, 09-2025· LMArena data1,197.8
97GPT 5.4 NanoHigh· LMArena data1,195.3
98GPT-4.5 Preview2025-02-27· LMArena data1,194.8
99o4 Mini2025-04-16· LMArena data1,193.9
100Claude Opus 4Thinking 16K, 20250514· LMArena data1,191.8

Showing 51–100 of 168 results.

Data credits: LMArena, CC BY 4.0 (licence)