Skip to main content
Radar

Provider changes are easy to miss. Radar watches releases, pricing, deprecations, and incidents at the source.Provider changes are easy to miss.

See Radar
BenchLM
Advertise

Benchmark profile

Korean Benchmark for Advanced Linguistic Tasks (KoBALT)

Evaluates advanced Korean linguistic competence.

Data verified 23 confirmed releases in the last 30 daysSee provider release alerts

How BenchLM shows KoBALT right now

BenchLM is tracking KoBALT in the local dataset, but exact-source verification records for these rows are still being attached. To avoid a blank benchmark page, BenchLM shows the current tracked rows below as a display-only reference table.

These tracked rows are useful for inspection and spot-checking, but until exact-source attachments are completed they should not be treated as fully verified public benchmark rows.

1 tracked modelsLocal tracked rowsAwaiting exact-source attachmentsDisplay only

Tracked score on KoBALT — August 7, 2026

BenchLM mirrors the published tracked score view for KoBALT. K-Exaone leads the public snapshot at 61.8%. BenchLM does not use these results to rank models overall.

1

K-Exaone

LG AI Research

k-exaone

61.8%
Overall —
1 modelKorean BenchmarksKorean-language benchmarkRefreshingDisplay onlyUpdated August 7, 2026

Tracked score table (1 model)

Score
1
K-ExaoneLG AI Research
61.8%

About KoBALT

Tasks

Linguistics questions

Format

Advanced linguistics

Difficulty

Advanced linguistic phenomena

BenchLM freshness & provenance

Version

KoBALT

Refresh cadence

Static

Staleness state

Refreshing

Question availability

Public benchmark set

RefreshingDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

FAQ

What does KoBALT measure?

Evaluates advanced Korean linguistic competence.

Which model leads the published KoBALT snapshot?

K-Exaone currently leads the published KoBALT snapshot with 61.8% tracked score. BenchLM shows this benchmark for display only and does not use it in overall rankings.

How many models are evaluated on KoBALT?

1 AI models are included in BenchLM's mirrored KoBALT snapshot, based on the public leaderboard captured on August 7, 2026.

Last updated: August 7, 2026 · mirrored from the public benchmark leaderboard

Know when it’s worth switching models

The model to choose, the cheaper alternative, and the release we would wait on.

Read a sample issue

Join 2,000+ readers.

One email each week. Unsubscribe anytime.