# KoBALT leaderboard

korean benchmark

> Every figure below is reproduced as its upstream source published it: nothing is modelled, estimated, interpolated or converted. `Unavailable` means no source published the value — it is never a zero. Each value carries its evidence state and the date it was observed.

Page: https://modelscale.dev/leaderboards/benchmarks/korean-kobalt

Korean Benchmark for Advanced Linguistic Tasks. Every model the catalog carries a published KoBALT value for, ranked by that value.

## Benchmark record

| Field | Value |
| --- | --- |
| Category | korean |
| Measure | Advanced linguistics |
| Tasks | Linguistics questions |
| Difficulty | Advanced linguistic phenomena |
| Published by | [BenchLM](https://benchlm.ai/benchmarks/kobalt) |

Evaluates advanced Korean linguistic competence.

## KoBALT ranking

2 models with a published KoBALT 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 | Advanced linguistics | Evidence |
| --- | --- | --- | --- | --- |
| 1 | [A.X K2](https://modelscale.dev/models/a-x-k2) | SK Telecom | 73 | Observed 2026-11-21 · source benchlm:models |
| 2 | [K-Exaone](https://modelscale.dev/models/k-exaone) | LG AI Research | 61.8 | Observed 2026-11-21 · source benchlm:models |

Rows are ordered by the value the source published, highest first. The source does not state whether a higher value is the better result, so this document does not either: for a benchmark that measures a rate of failure, read the table from the bottom.
