Skip to content

Benchmark

Arena ELO Text Factuality - Expert

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

Models measured
114
Scores recorded
130
Sources
1
IQ weight
2.21

Score distribution

1,480.4
Mean
1,482
Median
33.7
Std dev
1,437.1
10th percentile
1,522
90th percentile
1,386.2
Min
1,386.21,552.1
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.21 for chat models, measured over 114 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 Opus 5.5High· LMArena data1,552.1
2Claude Opus 5Max· LMArena data1,543.6
3Gemini 4 ArgonHigh· LMArena data1,541.9
4Claude Opus 4.6High· LMArena data1,541.6
5Claude Opus 4.6· LMArena data1,537.1
6GPT-6.1 SolMax· LMArena data1,535
7Claude Fable 5High· LMArena data1,534.7
8Claude Opus 5High· LMArena data1,533.4
9GPT 5.5High· LMArena data1,531.6
10GPT-5.6 SolxHigh· LMArena data1,529.6
11GPT 5.4High· LMArena data1,529.6
12Claude Fable 5.1Max· LMArena data1,528.5
13GPT 5.5· LMArena data1,523.2
14Claude Sonnet 5.5xHigh· LMArena data1,522
15DeepSeek V4.1 Flash 0910Max· LMArena data1,521.5
16Claude Opus 4.7· LMArena data1,521
17Claude Opus 4.7High· LMArena data1,520.4
18Muse WonderMax· LMArena data1,519.6
19Kimi K3Max· LMArena data1,517.9
20GPT-5.6 TerraxHigh· LMArena data1,516.3
21Claude Opus 4.8High· LMArena data1,515.9
22GPT-6 AstraMax· LMArena data1,515
23Gemini 3.7 FlashHigh· LMArena data1,513
24Muse Spark 1.3Max· LMArena data1,512.9
25GLM 5.3Max· LMArena data1,512.8
26Gemini 3.8 FlashHigh· LMArena data1,511.2
27GLM 5.3 Flash· LMArena data1,511.1
28MiMo V2.5 Pro· LMArena data1,510.7
29Claude Sonnet 4.6· LMArena data1,510.3
30GPT 5.4· LMArena data1,509.4
31Claude Opus 4.520251101· LMArena data1,507.6
32GPT-5.6 LunaxHigh· LMArena data1,505.5
33Gemini 3.5 FlashHigh· LMArena data1,505
34Qwen3.5-Max-Preview· LMArena data1,505
35ERNIE 5.1· LMArena data1,503.9
36Qwen3.8 Max· LMArena data1,503.2
37Claude Opus 4.5High, 32K, 20251101· LMArena data1,502.9
38Claude Opus 4.8· LMArena data1,502.5
39Kimi K2.6· LMArena data1,500.2
40MiMo V2.6 Flash· LMArena data1,499.7
41Step 5 Preview· LMArena data1,499.6
42GPT-6 LunaMax· LMArena data1,499.5
43Muse Spark 1.2xHigh· LMArena data1,497.5
44Gemini 3.6 FlashHigh· LMArena data1,497.3
45Claude Sonnet 5High· LMArena data1,497
46Gemini 3.5 FlashMedium· LMArena data1,496.6
47Gemini 3 Pro Preview· LMArena data1,495.9
48GLM 5.2Max· LMArena data1,495.3
49Grok 4.5· LMArena data1,495.1
50Claude Sonnet 4.5High, 32K, 20250929· LMArena data1,494.6

Showing 1–50 of 130 results.

Data credits: LMArena, CC BY 4.0 (licence)