Skip to content

Benchmark

Arena ELO Text Factuality - Industry Business and Management and Financial Operations

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

Models measured
146
Scores recorded
165
Sources
1
IQ weight
2.32

Score distribution

1,444.9
Mean
1,448.8
Median
35.8
Std dev
1,391.7
10th percentile
1,485.5
90th percentile
1,297.2
Min
1,297.21,509.6
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.32 for chat models, measured over 146 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 ArgonHigh· LMArena data1,509.6
2Kimi K3.1 ProMax· LMArena data1,504.4
3Muse WonderMax· LMArena data1,502.6
4Claude Fable 5.1Max· LMArena data1,501.5
5GPT 5.4High· LMArena data1,500
6Claude Opus 4.6High· LMArena data1,499.7
7GPT 5.5High· LMArena data1,499.2
8Claude Opus 5.5High· LMArena data1,499.2
9Gemini 4 NeonHigh· LMArena data1,497.9
10Claude Opus 5Max· LMArena data1,497.5
11GPT 5.5· LMArena data1,495.6
12Claude Opus 4.6· LMArena data1,492.9
13Claude Fable 5High· LMArena data1,491.6
14Claude Opus 4.7High· LMArena data1,489
15Muse Spark 1.3Max· LMArena data1,488.5
16Claude Opus 5High· LMArena data1,487.4
17Muse Spark 1.2xHigh· LMArena data1,485.5
18GPT 5.4· LMArena data1,485.2
19Qwen3.5-Max-Preview· LMArena data1,484.4
20Claude Opus 4.7· LMArena data1,484.3
21DeepSeek V4.1 Flash 0910Max· LMArena data1,484.3
22Qwen3.8 Max· LMArena data1,483.7
23Gemini 3.7 FlashHigh· LMArena data1,482.8
24GPT-5.6 SolxHigh· LMArena data1,481.6
25Kimi K3Max· LMArena data1,480.4
26Claude Opus 4.8High· LMArena data1,479.2
27Gemini 3.5 FlashHigh· LMArena data1,478.3
28Gemini 3.6 FlashHigh· LMArena data1,476.5
29MiMo V2.6 Pro· LMArena data1,476.2
30Gemini 3.8 FlashHigh· LMArena data1,476.1
31Claude Sonnet 5.5xHigh· LMArena data1,475.9
32GPT-5.6 TerraxHigh· LMArena data1,475.1
33Gemini 3 Pro Preview· LMArena data1,475.1
34MiMo V2.5 Pro· LMArena data1,474.5
35GPT-6 AstraMax· LMArena data1,473.3
36ERNIE 5.1· LMArena data1,472.6
37GPT-6.1 SolMax· LMArena data1,472.4
38Claude Opus 4.8· LMArena data1,472.2
39Claude Sonnet 4.6· LMArena data1,471.6
40GLM 5.3Max· LMArena data1,470.8
41Gemini 3.1 Pro Preview· LMArena data1,469.7
42Claude Fable 5.2Max· LMArena data1,469.4
43GPT-5.6 LunaxHigh· LMArena data1,467.6
44Muse Spark· LMArena data1,467.6
45Grok 4.5· LMArena data1,467.3
46Claude Opus 4.520251101· LMArena data1,466.9
47GLM 5.3 Flash· LMArena data1,466.6
48Gemini 3.5 FlashMedium· LMArena data1,465.5
49GPT 5.1High· LMArena data1,465.1
50Muse Spark 1.1· LMArena data1,465

Showing 1–50 of 165 results.

Data credits: LMArena, CC BY 4.0 (licence)