Skip to content

Benchmark

Arena ELO Vision - Homework

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

Models measured
104
Scores recorded
119
Sources
1
IQ weight
2.12

Score distribution

1,265.7
Mean
1,276
Median
57.7
Std dev
1,171.2
10th percentile
1,332.9
90th percentile
1,117.1
Min
1,117.11,362.8
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.12 for chat models, measured over 104 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
51GPT-5.4 miniHigh· LMArena data1,285.1
52Muse Spark 1.1· LMArena data1,284.1
53Gemini 2.5 Flash Preview09-2025· LMArena data1,283.6
54MiMo V2.6 Flash· LMArena data1,282.8
55GPT-5.6 LunaxHigh· LMArena data1,282.7
56Doubao Seed 2.0 Pro· LMArena data1,282.5
57Gemini 2.5 Pro· LMArena data1,281.5
58GPT-5 Chat· LMArena data1,279
59Qwen3 VL 235B· LMArena data1,278.9
60GPT 5.2High· LMArena data1,276
61Grok 4.20 Multi-Agent0309, beta· LMArena data1,275.6
62Step 5 Preview· LMArena data1,275.4
63Grok 4.6High· LMArena data1,273.8
64GPT 5.1High· LMArena data1,273.3
65Gemini 3.1 Flash Lite· LMArena data1,270.4
66Grok 4.20Reasoning, 0309, beta· LMArena data1,269.6
67MiniMax M3· LMArena data1,266.5
68Mistral Large 4· LMArena data1,265.5
69MiMo V2.6 Pro· LMArena data1,264.5
70Kimi K2.5Instant· LMArena data1,262.3
71Qwen3.5 122B-A10B· LMArena data1,260.7
72MiMo V2.5· LMArena data1,260.7
73GPT 5.2· LMArena data1,258.1
74GPT-6 LunaMax· LMArena data1,256.6
75ERNIE-5.0-Preview1220· LMArena data1,255.2
76Qwen3.5 27B· LMArena data1,254.4
77MiMo V2 Omni· LMArena data1,253.5
78ChatGPT-4o Latest20250326· LMArena data1,253.3
79Inkling Small· LMArena data1,251.3
80GPT 4.12025-04-14· LMArena data1,250.2
81GPT-5.5 Instant· LMArena data1,247.6
82Gemini 2.5 Flash· LMArena data1,244.9
83GPT 5.1· LMArena data1,244.6
84Grok 4.3· LMArena data1,241
85Hunyuan Vision 1.5 Thinking· LMArena data1,240.4
86Qwen3 VL 235Bthinking· LMArena data1,235.8
87OpenAI o32025-04-16· LMArena data1,235
88o4 Mini2025-04-16· LMArena data1,234.9
89NVIDIA Nemotron 3.5 Super VL 120BNVFP4· LMArena data1,230.6
90Mistral Medium 3.5· LMArena data1,229.2
91Claude Opus 420250514· LMArena data1,228
92GPT 5High· LMArena data1,226.5
93GPT 5.4 NanoHigh· LMArena data1,225.1
94Claude Sonnet 420250514· LMArena data1,215.6
95Gemini 2.5 Flash-Lite PreviewNon-Thinking, 09-2025· LMArena data1,215.2
96GPT-4.1 mini2025-04-14· LMArena data1,213.8
97GPT-5 miniHigh· LMArena data1,212.4
98Mistral Large 3· LMArena data1,212.3
99Claude Sonnet 3.720250219· LMArena data1,206.8
100Claude Opus 4Thinking 16K, 20250514· LMArena data1,197.6

Showing 51–100 of 119 results.

Data credits: LMArena, CC BY 4.0 (licence)