Skip to content

Benchmark

Arena ELO Text Factuality - Non English

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

Models measured
168
Scores recorded
192
Sources
1
IQ weight
2.36

Score distribution

1,422.4
Mean
1,430.7
Median
42.6
Std dev
1,361
10th percentile
1,469
90th percentile
1,245.6
Min
1,245.61,512.3
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.36 for chat models, measured over 168 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 data1,512.3
2Gemini 4 NeonHigh· LMArena data1,505.7
3Claude Opus 5.5High· LMArena data1,492.1
4Claude Fable 5.1Max· LMArena data1,489.4
5Gemini 4 HydrogenHigh· LMArena data1,486.5
6Gemini 4 ArgonHigh· LMArena data1,483.5
7Muse WonderMax· LMArena data1,482.1
8Claude Opus 5Max· LMArena data1,479.5
9Claude Opus 4.6High· LMArena data1,478.6
10GPT 5.5High· LMArena data1,478.2
11Claude Fable 5.2Max· LMArena data1,476
12Qwen3.7 Max Preview· LMArena data1,475.4
13GPT 5.4High· LMArena data1,475.3
14Claude Opus 4.6· LMArena data1,475.2
15Gemini 3 Pro Preview· LMArena data1,474.2
16Claude Fable 5High· LMArena data1,471.5
17GPT 5.5· LMArena data1,470.7
18Gemini 3.5 FlashHigh· LMArena data1,469.8
19Muse Spark 1.3Max· LMArena data1,469.6
20Gemini 3.7 FlashHigh· LMArena data1,469
21Claude Opus 4.7High· LMArena data1,468.1
22Claude Opus 5High· LMArena data1,467.6
23Qwen3.5-Max-Preview· LMArena data1,467.1
24GPT-5.6 SolxHigh· LMArena data1,466.7
25MiMo V2.6 Pro· LMArena data1,466.6
26Claude Opus 4.7· LMArena data1,466.3
27Kimi K3.1 ProMax· LMArena data1,466.2
28Muse Spark 1.2xHigh· LMArena data1,465.8
29GPT 5.4· LMArena data1,465.1
30Gemini 3.1 Pro Preview· LMArena data1,463.3
31Gemini 3.8 FlashHigh· LMArena data1,462.8
32Gemini 3.6 FlashHigh· LMArena data1,462.4
33Qwen3.8 Max· LMArena data1,460.3
34GPT-6 LunaMax· LMArena data1,459.5
35Gemini 3 Flash· LMArena data1,459.1
36GPT-6.1 SolMax· LMArena data1,458.6
37GPT-5.6 TerraxHigh· LMArena data1,458.4
38Claude Opus 4.8High· LMArena data1,458.4
39Gemini 3.5 FlashMedium· LMArena data1,458.3
40Kimi K3Max· LMArena data1,458.1
41Claude Sonnet 5.5xHigh· LMArena data1,457.1
42GPT-6 AstraMax· LMArena data1,456.7
43Muse Spark· LMArena data1,456.6
44GLM 5.3Max· LMArena data1,456
45Grok 4.5· LMArena data1,453.8
46DeepSeek V4.1 Flash 0910Max· LMArena data1,453.6
47Claude Opus 4.8· LMArena data1,453
48Muse Spark 1.1· LMArena data1,453
49Gemini 2.5 Pro· LMArena data1,453
50GLM 5.3 Flash· LMArena data1,449.6

Showing 1–50 of 192 results.

Data credits: LMArena, CC BY 4.0 (licence)