Skip to content

Benchmark

Arena ELO Text Factuality - Industry Life and Physical and Social Science

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

Models measured
138
Scores recorded
156
Sources
1
IQ weight
2.35

Score distribution

1,466.8
Mean
1,468.6
Median
33.7
Std dev
1,433.2
10th percentile
1,506.4
90th percentile
1,348.5
Min
1,348.51,551.3
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.35 for chat models, measured over 138 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 NeonHigh· LMArena data1,551.3
2Claude Fable 5.1Max· LMArena data1,533.1
3Gemini 4 ArgonHigh· LMArena data1,532
4Claude Opus 5.5High· LMArena data1,530.5
5Claude Opus 5Max· LMArena data1,528.9
6Claude Opus 4.6High· LMArena data1,521.6
7Gemini 4 HydrogenHigh· LMArena data1,514.5
8Claude Opus 5High· LMArena data1,513.9
9Muse WonderMax· LMArena data1,513.9
10Claude Opus 4.6· LMArena data1,512.6
11Claude Fable 5High· LMArena data1,512.5
12Qwen3.5-Max-Preview· LMArena data1,512
13Qwen3.8 Max· LMArena data1,511.3
14Gemini 3.7 FlashHigh· LMArena data1,511.1
15Claude Fable 5.2Max· LMArena data1,509.2
16GPT 5.5· LMArena data1,506.4
17GPT 5.5High· LMArena data1,505.9
18Claude Opus 4.7High· LMArena data1,505.7
19Gemini 3.5 FlashHigh· LMArena data1,503.1
20Gemini 3.8 FlashHigh· LMArena data1,502.9
21Muse Spark 1.3Max· LMArena data1,502.8
22Muse Spark 1.2xHigh· LMArena data1,501.6
23Kimi K3Max· LMArena data1,501.5
24Claude Opus 4.7· LMArena data1,500.7
25Kimi K3.1 ProMax· LMArena data1,500.5
26Gemini 3 Pro Preview· LMArena data1,499.9
27GPT 5.4High· LMArena data1,499.3
28GLM 5.3Max· LMArena data1,499.2
29Claude Opus 4.8High· LMArena data1,497.4
30Gemini 3.5 FlashMedium· LMArena data1,495.1
31Gemini 3.1 Pro Preview· LMArena data1,494.8
32Gemini 3.6 FlashHigh· LMArena data1,493.7
33Claude Sonnet 5.5xHigh· LMArena data1,493.4
34GPT-6.1 SolMax· LMArena data1,492.9
35GLM 5.2Max· LMArena data1,492.8
36DeepSeek V4.1 Flash 0910Max· LMArena data1,491.4
37Gemini 2.5 Pro· LMArena data1,491
38MiMo V2.5 Pro· LMArena data1,490
39GPT-6 AstraMax· LMArena data1,489.7
40Claude Opus 4.8· LMArena data1,489.3
41GPT 5.4· LMArena data1,488.7
42Grok 4.5· LMArena data1,488.4
43GPT-5.6 SolxHigh· LMArena data1,488.4
44GLM 5.3 Flash· LMArena data1,486.8
45Claude Sonnet 4.6· LMArena data1,486.3
46GLM 5.1· LMArena data1,485.5
47Qwen3 Max Preview· LMArena data1,485.3
48Muse Spark· LMArena data1,485.1
49Gemini 3 Flash· LMArena data1,484.8
50GPT-5.6 TerraxHigh· LMArena data1,483.7

Showing 1–50 of 156 results.

Data credits: LMArena, CC BY 4.0 (licence)