Skip to content

Benchmark

Arena ELO Vision - Humor

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

Models measured
90
Scores recorded
102
Sources
1
IQ weight
1.76

Score distribution

1,244.3
Mean
1,252.6
Median
54.3
Std dev
1,167.5
10th percentile
1,313
90th percentile
1,100.3
Min
1,100.31,347.9
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.76 for chat models, measured over 90 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Claude Fable 5High· LMArena data1,347.9
2Muse Spark· LMArena data1,343.2
3Gemini 3 Pro Preview· LMArena data1,337.3
4Claude Fable 5.1Max· LMArena data1,324.1
5Gemini 3.1 Pro Preview· LMArena data1,323.8
6Gemini 3.5 FlashMedium· LMArena data1,322.4
7Gemini 3.5 FlashHigh· LMArena data1,318.8
8Claude Opus 4.6· LMArena data1,315.9
9Claude Opus 4.7High· LMArena data1,314
10Claude Opus 4.6High· LMArena data1,313.2
11Claude Opus 4.7· LMArena data1,313
12Gemini 3.8 FlashHigh· LMArena data1,310.1
13Qwen3.8 Max· LMArena data1,304.9
14Gemini 3.7 FlashHigh· LMArena data1,302.3
15GPT 5.5High· LMArena data1,297.1
16Muse Spark 1.1· LMArena data1,296
17Claude Opus 5.5High· LMArena data1,294.7
18Claude Opus 5High· LMArena data1,292.6
19Gemini 3 Flash· LMArena data1,292
20GLM 5.3 Flash· LMArena data1,291.1
21Gemini 3.6 FlashHigh· LMArena data1,285.6
22GPT 5.4High· LMArena data1,285.1
23Kimi K2.5Thinking· LMArena data1,285
24Claude Opus 4.8· LMArena data1,283.3
25Grok 4.20 Multi-Agent0309, beta· LMArena data1,282.4
26Claude Opus 4.8High· LMArena data1,281.3
27Grok 4.20Reasoning, 0309, beta· LMArena data1,280.1
28Gemma 4 31B IT· LMArena data1,280
29Kimi K2.6· LMArena data1,279.8
30Muse Spark 1.3Max· LMArena data1,279
31Gemini 3 FlashMinimal· LMArena data1,278.1
32GPT 5.5· LMArena data1,277.3
33Gemma 4 26B A4B· LMArena data1,275.7
34Doubao Seed 2.0 Pro· LMArena data1,275.5
35Grok 4.5· LMArena data1,274.4
36Gemini 3.5 Flash Lite· LMArena data1,273.3
37DeepSeek V4.1 Flash 0910Max· LMArena data1,272
38GPT-6.1 SolMax· LMArena data1,271.8
39Claude Sonnet 5.5xHigh· LMArena data1,271
40GPT-6 AstraMax· LMArena data1,270.6
41Grok 4.6High· LMArena data1,267.9
42GPT-5.5 Instant· LMArena data1,267.7
43Gemini 3.1 Flash Lite· LMArena data1,266.8
44Muse Spark 1.3xHigh· LMArena data1,266.6
45Qwen3.7 Plus· LMArena data1,265.9
46GPT 5.4· LMArena data1,264.4
47Qwen3.5 397B A17B· LMArena data1,264.1
48GPT 5.1· LMArena data1,263.3
49Gemini 2.5 Pro· LMArena data1,261.6
50Claude Sonnet 5High· LMArena data1,255.2

Showing 1–50 of 102 results.

Data credits: LMArena, CC BY 4.0 (licence)