Skip to content

Benchmark

Arena ELO Text Factuality - Korean

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

Models measured
46
Scores recorded
52
Sources
1
IQ weight
2.08

Score distribution

1,408.1
Mean
1,410.2
Median
34.3
Std dev
1,367.4
10th percentile
1,454.1
90th percentile
1,325.5
Min
1,325.51,498.1
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.08 for chat models, measured over 46 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 3.8 FlashHigh· LMArena data1,498.1
2Claude Opus 5High· LMArena data1,472.5
3Claude Opus 4.7· LMArena data1,462.4
4Claude Opus 4.6· LMArena data1,461.9
5Gemini 3.1 Pro Preview· LMArena data1,460.4
6GPT 5.5High· LMArena data1,454.1
7Gemini 3 Pro Preview· LMArena data1,451.3
8GPT-5.6 SolxHigh· LMArena data1,448.4
9GPT 5.5· LMArena data1,447.5
10GPT 5.4· LMArena data1,438.7
11Claude Opus 4.6High· LMArena data1,438.5
12GPT-5.6 TerraxHigh· LMArena data1,434.3
13Claude Opus 4.7High· LMArena data1,432.9
14GLM 5.2Max· LMArena data1,430.1
15Gemini 3 FlashMinimal· LMArena data1,422.6
16Grok 4.20Reasoning, 0309, beta· LMArena data1,422.3
17Kimi K2.5Thinking· LMArena data1,417.5
18Claude Opus 4.8· LMArena data1,417.2
19Claude Opus 4.520251101· LMArena data1,416.5
20Gemini 2.5 Pro· LMArena data1,415.7
21DeepSeek V4 Pro· LMArena data1,414.4
22MiMo V2.5 Pro· LMArena data1,414.4
23GLM 5.1· LMArena data1,414.3
24Kimi K3Max· LMArena data1,412.8
25Doubao Seed 2.0 Pro· LMArena data1,412
26Claude Sonnet 4.520250929· LMArena data1,410.2
27Claude Sonnet 4.6· LMArena data1,407.3
28Grok 4.1· LMArena data1,406.9
29Gemini 3.1 Flash Lite· LMArena data1,404.4
30Qwen3 235B A22B Instruct 25072507· LMArena data1,403.9
31GPT 5.2· LMArena data1,401.4
32GLM 5.3 Flash· LMArena data1,400.1
33Qwen3.5 397B A17B· LMArena data1,399.3
34Gemini 2.5 Flash· LMArena data1,395.6
35GPT-6 LunaMax· LMArena data1,392.5
36GPT-5.4 miniHigh· LMArena data1,390.2
37Grok 4.1Thinking· LMArena data1,383.7
38Mistral Large 3· LMArena data1,381.9
39MiniMax M3· LMArena data1,381.1
40MiMo V2.5· LMArena data1,378.2
41Claude Sonnet 4.5High, 32K, 20250929· LMArena data1,378.1
42Grok 4.1 FastReasoning· LMArena data1,376.3
43GPT 5.2High· LMArena data1,374.3
44Qwen3.6 Plus· LMArena data1,372.4
45Grok 4.3· LMArena data1,369.7
46Claude Haiku 4.520251001· LMArena data1,368.1
47Mistral Medium 3.12508· LMArena data1,367.4
48Qwen3.5 Flash· LMArena data1,367.2
49Step 3.5 Flash· LMArena data1,366.7
50DeepSeek-V4-Flash PreviewHigh· LMArena data1,361.3

Showing 1–50 of 52 results.

Data credits: LMArena, CC BY 4.0 (licence)