Skip to content

Benchmark

Arena ELO Text Factuality - Industry Writing and Literature and Language

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

Models measured
154
Scores recorded
173
Sources
1
IQ weight
2.19

Score distribution

1,424.1
Mean
1,427.8
Median
44.3
Std dev
1,368.5
10th percentile
1,472.9
90th percentile
1,271.6
Min
1,271.61,520.8
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.19 for chat models, measured over 154 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Gemini 4 CarbonHigh· LMArena data1,520.8
2Claude Opus 5.5High· LMArena data1,511.8
3Gemini 4 ArgonHigh· LMArena data1,503.5
4Claude Fable 5.1Max· LMArena data1,503.3
5Gemini 4 NeonHigh· LMArena data1,498.8
6Claude Fable 5.2Max· LMArena data1,493.7
7Claude Fable 5High· LMArena data1,489.9
8Claude Opus 4.6High· LMArena data1,489.5
9Gemini 4 HydrogenHigh· LMArena data1,486.7
10Muse WonderMax· LMArena data1,485.8
11Claude Opus 5Max· LMArena data1,485.1
12Claude Opus 4.6· LMArena data1,481.3
13Claude Opus 4.7High· LMArena data1,481
14Gemini 3.5 FlashHigh· LMArena data1,477.7
15Gemini 3.7 FlashHigh· LMArena data1,476.4
16Gemini 3 Pro Preview· LMArena data1,475.8
17Claude Opus 5High· LMArena data1,474.8
18Qwen3.8 Max· LMArena data1,472.9
19Claude Opus 4.7· LMArena data1,472
20Gemini 3.8 FlashHigh· LMArena data1,471.8
21Gemini 3.1 Pro Preview· LMArena data1,470.4
22GPT 5.5High· LMArena data1,469.8
23Qwen3.5-Max-Preview· LMArena data1,469.6
24Kimi K3.1 ProMax· LMArena data1,469.5
25Muse Spark 1.3Max· LMArena data1,469.3
26GPT 5.5· LMArena data1,469.3
27Claude Sonnet 5.5xHigh· LMArena data1,468.9
28GPT-5.6 SolxHigh· LMArena data1,468.8
29Claude Opus 4.8High· LMArena data1,466.9
30GPT 5.4High· LMArena data1,464
31Gemini 3.6 FlashHigh· LMArena data1,463.7
32Muse Spark· LMArena data1,463.7
33GLM 5.3Max· LMArena data1,463.7
34Kimi K3Max· LMArena data1,463.6
35Gemini 3.5 FlashMedium· LMArena data1,463.6
36Gemini 3 Flash· LMArena data1,463.2
37MiMo V2.6 Pro· LMArena data1,460.9
38GPT 5.4· LMArena data1,460
39GLM 5.2Max· LMArena data1,458.8
40Muse Spark 1.2xHigh· LMArena data1,456.5
41Claude Opus 4.8· LMArena data1,456.2
42Grok 4.5· LMArena data1,454.7
43GPT-6.1 SolMax· LMArena data1,454.3
44Claude Opus 4.520251101· LMArena data1,454.2
45GPT-6 AstraMax· LMArena data1,453.2
46MiMo V2.5 Pro· LMArena data1,452.9
47GLM 5.1· LMArena data1,452.7
48DeepSeek V4.1 Flash 0910Max· LMArena data1,452
49Claude Opus 4.5High, 32K, 20251101· LMArena data1,451.5
50Claude Sonnet 4.520250929· LMArena data1,451.1

Showing 1–50 of 173 results.

Data credits: LMArena, CC BY 4.0 (licence)