Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
korean benchmark
Korean Massive Multitask Language Understanding. Every model the catalog carries a published KMMLU value for, ranked by that value.
Evaluates Korean expert-level knowledge across 45 subjects. 20% of questions require Korean cultural context. 2 models in the ModelScale catalog have a published value.
Each point is a model with a published KMMLU value and complete token prices, on the 75% input / 25% output blend. Frontier models are spaced evenly along the cost axis.
2 models with a published KMMLU value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.
| Rank | Model | Provider | Multiple choice questions |
|---|---|---|---|
| 1 | KAKanana-2 3B Instruct | Kakao | 43.32 |
| 2 | KAKanana-2 1.3B Instruct | Kakao | 42.79 |
Evidence key: Observed
Rows are ordered by the value KMMLU: Measuring Massive Multitask Language Understanding in Korean published, highest first. The source does not state whether a higher value is the better result, so this page does not either: for a benchmark that measures a rate of failure, read the table from the bottom.
What the benchmark measures and how its value is published, from the source’s own record.
KMMLU (Korean Massive Multitask Language Understanding) is a korean benchmark published by KMMLU: Measuring Massive Multitask Language Understanding in Korean. Evaluates Korean expert-level knowledge across 45 subjects. 20% of questions require Korean cultural context. Its evaluation set is described by the source as 35,030 questions. The source rates its difficulty as elementary to professional level in Korean.
Each model's result is published by KMMLU: Measuring Massive Multitask Language Understanding in Korean as multiple choice questions. ModelScale prints that value exactly as published, and never converts or rescales it. This page ranks the 2 models that carry a published value, highest value first; a model the source never scored is left out rather than ranked at zero. The source does not state whether a higher value is the better result, so neither does this page.