Skip to content

Benchmark

Arena ELO - Korean

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

Models measured
244
Scores recorded
288
Sources
1
IQ weight
1.99

Score distribution

1,298.9
Mean
1,348.8
Median
139.4
Std dev
1,087.1
10th percentile
1,441.2
90th percentile
856
Min
8561,509.5
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.99 for chat models, measured over 244 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 data1,509.5
2Gemini 4 ArgonHigh· LMArena data1,505.5
3Claude Opus 5Max· LMArena data1,503.1
4Claude Opus 5High· LMArena data1,502
5Claude Fable 5.1Max· LMArena data1,499
6Muse Spark 1.3Max· LMArena data1,484.8
7Gemini 3.7 FlashHigh· LMArena data1,483.3
8Claude Fable 5High· LMArena data1,481.7
9Gemini 3.8 FlashHigh· LMArena data1,473.8
10Muse WonderMax· LMArena data1,464.6
11Claude Opus 4.7· LMArena data1,464.5
12Muse Spark· LMArena data1,461.5
13Claude Opus 4.6· LMArena data1,459.9
14Qwen3.8 Max· LMArena data1,457.7
15Gemini 3.1 Pro Preview· LMArena data1,457.2
16Gemini 3.5 FlashMedium· LMArena data1,456.6
17Muse Spark 1.1· LMArena data1,456.1
18Claude Opus 4.6High· LMArena data1,452.4
19GPT 5.5High· LMArena data1,452.1
20Gemini 3.5 FlashHigh· LMArena data1,449.6
21GPT-6.1 SolMax· LMArena data1,449.5
22Claude Opus 4.7High· LMArena data1,448.9
23GPT 5.4High· LMArena data1,448.6
24GPT-5.6 SolxHigh· LMArena data1,447.5
25Gemini 3 Flash· LMArena data1,447.4
26Gemini 3 Pro Preview· LMArena data1,447.3
27Muse Spark 1.2xHigh· LMArena data1,445.1
28Gemini 3.6 FlashHigh· LMArena data1,442.6
29GLM 5.3Max· LMArena data1,441.6
30Kimi K3Max· LMArena data1,441.2
31GPT 5.5· LMArena data1,441
32MiMo V2.5 Pro· LMArena data1,439.4
33Qwen3.5-Max-Preview· LMArena data1,438.9
34GPT 5.4· LMArena data1,438.1
35Gemini 2.5 Pro· LMArena data1,433.7
36DeepSeek-V4-Pro PreviewHigh· LMArena data1,430.7
37GLM 5.2Max· LMArena data1,428.7
38Kimi K2.6· LMArena data1,428.5
39ERNIE 5.1· LMArena data1,427.9
40GLM 5.3 Flash· LMArena data1,426.8
41Claude Opus 4.520251101· LMArena data1,425.5
42GLM 5.1· LMArena data1,424.5
43GPT-5.6 TerraxHigh· LMArena data1,424.4
44Claude Opus 4.8High· LMArena data1,423.6
45DeepSeek V4.1 Flash 0910Max· LMArena data1,423
46Claude Sonnet 5.5xHigh· LMArena data1,422.8
47Grok 4.20Reasoning, 0309, beta· LMArena data1,421.4
48Gemini 3 FlashMinimal· LMArena data1,420.3
49GLM 5· LMArena data1,419.7
50Doubao Seed 2.0 Pro· LMArena data1,419.6

Showing 1–50 of 288 results.

Data credits: LMArena, CC BY 4.0 (licence)