Valumigo

LMArena · Korean

We regularly fetch and display public benchmark data: LMArena user-voted rankings and Epoch AI test scores. Each leaderboard shows its publication and retrieval dates. These scores are not produced by this site.

LMArena scores come from people comparing two models' answers side by side and voting for the better one, using an Elo-based system. If score differences fall within the confidence intervals, treat the models as roughly comparable.

How to read this leaderboard

What it measures: A ranking calculated from votes on questions classified as Korean in LMArena text conversations.

How to read the score: Shows which models voters preferred for questions in Korean. The question's language does not indicate the voter's native language or identity.

Caveats: Vote counts and confidence interval widths vary by language and model. For new models, check vote counts, update dates and aggregation conditions together.

Insights from this ranking

  • The 95% confidence intervals for 1st-place claude-fable-5.1-max and 2nd-place claude-opus-5-max overlap. This aggregation alone does not clearly establish their order.
  • 9 other models have confidence intervals that overlap with the 1st-place model's. Interpret small ranking differences cautiously alongside vote counts.
  • Anthropic has the most models among the top 10, with 5.
  • The 1st-place model received 177 votes, and this leaderboard includes 278 models.

Korean LMArena

Published 2026-10-02 · Retrieved 2026-10-04

14001450150015501600
Anthropicclaude-fable-5.1-max
1534
Anthropicclaude-opus-5-max
1519
Metamuse-spark-1.3-max
1516
Anthropicclaude-opus-5-high
1511
Anthropicclaude-fable-5-high
1488
Googlegemini-3.7-flash-high
1483
Zglm-5.3-max
1475
Googlegemini-3.8-flash-high
1468
Alibabaqwen3.8-max
1465
Anthropicclaude-opus-4-6
1465
Anthropicclaude-opus-4-7
1464
Ogpt-5.5-high
1461
Metamuse-spark-1.1
1461
Moonshot AIkimi-k3-max
1460
Googlegemini-3.1-pro-preview
1458
Metamuse-spark
1457
Anthropicclaude-opus-4-7-high
1453
Googlegemini-3.5-flash-medium
1452
Ogpt-5.4-high
1450
Anthropicclaude-opus-4-6-high
1450

Dots show scores; horizontal lines show 95% confidence intervals. Overlapping intervals suggest similar performance.

View table (top 50)
RankModelDeveloperScore95% confidence intervalVotes
1Anthropicclaude-fable-5.1-maxAnthropic15341487–1581177
2Anthropicclaude-opus-5-maxAnthropic15191494–1545568
3Metamuse-spark-1.3-maxMeta15161475–1558205
4Anthropicclaude-opus-5-highAnthropic15111492–15311,126
5Anthropicclaude-fable-5-highAnthropic14881465–1511722
6Googlegemini-3.7-flash-highGoogle14831452–1514404
7Zglm-5.3-maxZ.ai14751441–1510298
8Googlegemini-3.8-flash-highGoogle14681440–1497487
9Alibabaqwen3.8-maxAlibaba14651438–1493493
10Anthropicclaude-opus-4-6Anthropic14651448–14821,486
11Anthropicclaude-opus-4-7Anthropic14641446–14831,144
12Ogpt-5.5-highOpenAI14611443–14791,187
13Metamuse-spark-1.1Meta14611438–1484708
14Moonshot AIkimi-k3-maxMoonshot AI14601436–1484615
15Googlegemini-3.1-pro-previewGoogle14581443–14732,145
16Metamuse-sparkMeta14571419–1496252
17Anthropicclaude-opus-4-7-highAnthropic14531434–14731,100
18Googlegemini-3.5-flash-mediumGoogle14521431–1473855
19Ogpt-5.4-highOpenAI14501430–14691,106
20Anthropicclaude-opus-4-6-highAnthropic14501432–14671,310
21Googlegemini-3-proGoogle14491426–1472719
22Googlegemini-3.5-flash-highGoogle14481429–1468960
23Zglm-5.2-maxZ.ai14471427–1468879
24Ogpt-5.5OpenAI14451427–14641,180
25Googlegemini-3-flashGoogle14431418–1469579
26DeepSeekdeepseek-v4-pro-high-20260813DeepSeek14431403–1484195
27Zglm-5.3-flashZ.ai14391408–1469397
28Ogpt-5.4OpenAI14381419–14571,159
29Alibabaqwen3.5-max-previewAlibaba14381409–1467411
30Googlegemini-3.6-flash-highGoogle14381414–1461685
31Ogpt-5.6-sol-xhighOpenAI14361413–1459723
32Xiaomimimo-v2.5-proXiaomi14351417–14531,305
33Googlegemini-2.5-proGoogle14341421–14472,476
34Anthropicclaude-opus-4-8-highAnthropic14331414–14531,094
35Moonshot AIkimi-k2.6Moonshot AI14291406–1452729
36Baiduernie-5.1Baidu14281404–1452654
37DeepSeekdeepseek-v4-pro-high-previewDeepSeek14261407–1446980
38Ogpt-5.6-terra-xhighOpenAI14251403–1448733
39Anthropicclaude-opus-4-5-20251101Anthropic14251407–14431,163
40Anthropicclaude-opus-4-8Anthropic14251405–14441,064
41Zglm-5Z.ai14231397–1448552
42Zglm-5.1Z.ai14201401–14391,113
43Googlegemini-3-flash (thinking-minimal)Google14181401–14351,408
44xgrok-4.20-multi-agent-beta-0309xAI14171397–14361,045
45xgrok-4.20-beta1xAI14161388–1444467
46xgrok-4.20-beta-0309-reasoningxAI14161396–14361,058
47Alibabaqwen3.7-plusAlibaba14161394–1437828
48Ogpt-5.6-luna-xhighOpenAI14141392–1437712
49ByteDancedola-seed-2.0-proByteDance14131395–14301,322
50Moonshot AIkimi-k2.5-thinkingMoonshot AI14121394–14291,355

Data: LMArena leaderboard dataset (CC BY 4.0). Changes: Top 50 entries per leaderboard; scores are rounded. https://huggingface.co/datasets/lmarena-ai/leaderboard-datasetCompany logos are trademarks of their respective owners and are used only for identification (icons: Simple Icons).Data: Epoch AI — Capabilities & benchmarking (CC BY 4.0). https://epoch.ai/benchmarks