GGearDel.
한국어

Collection

Korean LLM benchmark datasets

47 catalog items · GearDel

GearDel lists 47 catalog items in this group (46 text, 1 image). Sources: 27 Hugging Face, 3 GitHub. Examples: KLUE Benchmark, KMMLU, CLIcK (Cultural & Legal Knowledge). 35 are marked commercial-use-allowed in the catalog. These are evaluation sets, not pretraining dumps. Use them to score models, then train on a separate corpus. Use this list to compare license and size, then open the original provider. GearDel does not host files.

KLUE, KoBEST, and KMMLU-family rows are for leaderboards. If you need training data, leave this table for the LLM training collection. Training on test items raises a score, not generalization.

Clauses differ. If the catalog shows something like CC BY-ND, run the eval script and do not treat weights as redistributable training data.

Split tasks (NLI, NER, MRC) do not all need to be downloaded. Open the task you will score and keep the host’s official splits.

Small archives can still be carefully designed items. GearDel megabyte figures are archive hints, not difficulty scores.

Do not train on the test split. KLUE, KoBEST, and KMMLU families are for scoring, then go back to an instruction or corpus collection.

Browser tools that stay on this device

Allowed

35

Conditional

11

Restricted

1

Counts are rows in this collection, not downloads. License labels below are the stored strings.

  • CC BY-SA 4.0 · 14
  • MIT · 11
  • Apache 2.0 · 6
  • CC BY-NC 4.0 · 6
  • License unknown · 5
  • CC BY-NC-SA 4.0 · 2
  • CC BY 4.0 · 1
  • CC BY-NC-ND 4.0 · 1
NameUse / topicSourceSizeLicenseCommercial
KLUE BenchmarkKorean LLM Benchmark DatasetHugging Face62.3 MBCC BY-SA 4.0Commercial use may be conditional
KMMLUKorean LLM Benchmark DatasetHugging Face70.7 MBCC BY-ND 4.0Commercial use restricted
CLIcK (Cultural & Legal Knowledge)Korean LLM Benchmark DatasetGitHub5.5MBLicense unknownCommercial use allowed (catalog)
KoBESTKorean LLM Benchmark DatasetHugging Face5.1MBCC BY-SA 4.0Commercial use allowed (catalog)
KoBigBench LiteKorean LLM Benchmark DatasetHugging Face12MBLicense unknownCommercial use allowed (catalog)
KoInFoBenchKorean Instruction Tuning DatasetHugging FaceSmallMITCommercial use allowed (catalog)
Korean QA DatasetKorean QA DatasetHugging FaceSmallApache 2.0Commercial use allowed (catalog)
Korean LLM Benchmark DatasetKorean LLM Benchmark DatasetHugging Face (SOGANG-ISDS)SmallApache 2.0Commercial use allowed (catalog)
SNU Ko-MuSRKorean LLM Benchmark DatasetHugging Face (thunder-research-group)SmallMITCommercial use allowed (catalog)
KLUEKorean NLP DatasetHugging Face (KLUE)SmallCC BY-SA 4.0Commercial use allowed (catalog)
KLUEKorean NLP DatasetHugging Face (KLUE)SmallCC BY-SA 4.0Commercial use allowed (catalog)
KLUEKorean LLM Benchmark DatasetHugging Face (KLUE)SmallCC BY-SA 4.0Commercial use allowed (catalog)
KLUEKorean NER DatasetHugging Face (KLUE)MediumCC BY-SA 4.0Commercial use allowed (catalog)
KLUEKorean NLP DatasetHugging Face (KLUE)SmallCC BY-SA 4.0Commercial use allowed (catalog)
KLUEKorean QA DatasetHugging Face (KLUE)MediumCC BY-SA 4.0Commercial use allowed (catalog)
COPAKorean LLM Benchmark DatasetHugging Face (SKT-brain)SmallCC BY-SA 4.0Commercial use allowed (catalog)
HellaSwagKorean LLM Benchmark DatasetHugging Face (SKT-brain)SmallCC BY-SA 4.0Commercial use allowed (catalog)
BoolQKorean QA DatasetHugging Face (SKT-brain)SmallCC BY-SA 4.0Commercial use allowed (catalog)
WiCKorean NLP DatasetHugging Face (SKT-brain)SmallCC BY-SA 4.0Commercial use allowed (catalog)
Korean LLM Benchmark DatasetKorean LLM Benchmark DatasetGitHubSmallMITCommercial use allowed (catalog)
KMMLU-ProKorean LLM Benchmark DatasetHugging Face (LGAI-EXAONE)12MBCC BY-NC 4.0Commercial use allowed (catalog)
Thunder-KoNUBenchKorean LLM Benchmark DatasetHugging Face (thunder-research-group)5.4MBCC BY-NC-SA 4.0Commercial use allowed (catalog)
K-BrowseCompKorean LLM Benchmark DatasetHugging Face (prometheus-eval)3.5MBMITCommercial use allowed (catalog)
CSAT-KOREAN-2025Korean QA DatasetHugging Face (KKACHI-HUB)3.2MBMITCommercial use allowed (catalog)
Korean LLM Benchmark DatasetKorean LLM Benchmark DatasetHugging Face / GitHubMediumApache 2.0Commercial use allowed (catalog)
Korean LLM Benchmark DatasetKorean LLM Benchmark DatasetHugging FaceSmallApache 2.0Commercial use allowed (catalog)
KoGEMKorean LLM Benchmark DatasetHugging Face1.2MBMITCommercial use allowed (catalog)
Ko-MMLU-ProKorean LLM Benchmark DatasetHugging Face45MBMITCommercial use allowed (catalog)
KoSafetyBenchKorean LLM Benchmark DatasetHugging Face15MBCC BY 4.0Commercial use allowed (catalog)
Ko-ARC ChallengeKorean LLM Benchmark DatasetHugging Face8MBCC BY-SA 4.0Commercial use allowed (catalog)
HAE-RAE Bench 1.1Korean LLM Benchmark DatasetHugging Face2.9MBCC BY-NC-ND 4.0Commercial use may be conditional
HAE-RAE Bench 2.0Korean LLM Benchmark DatasetHugging Face1.0MBMITCommercial use allowed (catalog)
KoCommonGEN v2Korean LLM Benchmark DatasetHugging Face183KBLicense unknownCommercial use may be conditional
APEACHKorean LLM Benchmark DatasetHugging Face1.1MBCC BY-SA 4.0Commercial use allowed (catalog)
HRM8K (HAE-RAE Math 8K)Korean LLM Benchmark DatasetHugging Face4.9MBMITCommercial use allowed (catalog)
KorNAT QAKorean QA DatasetHugging Face2.6MBCC BY-NC 4.0Commercial use may be conditional
HRMCRKorean LLM Benchmark DatasetHugging Face106KBApache 2.0Commercial use allowed (catalog)
KoBBQ QAKorean QA DatasetHugging Face2.4MBMITCommercial use allowed (catalog)
KorMedMCQA QAKorean QA DatasetHugging Face2.1MBCC BY-NC 4.0Commercial use may be conditional
KoBALT-700Korean LLM Benchmark DatasetHugging Face644KBCC BY-NC 4.0Commercial use may be conditional
KoSimpleQA QAKorean QA DatasetHugging Face301KBLicense unknownCommercial use may be conditional
Ko-PIQAKorean QA DatasetHugging Face204KBLicense unknownCommercial use may be conditional
K-MMBenchKorean Computer Vision DatasetHugging Face92.4MBCC BY-NC 4.0Commercial use may be conditional
EvalKorean Instruction Tuning DatasetHugging Face178KBMITCommercial use allowed (catalog)
KoDialogBenchKorean LLM Benchmark DatasetHugging Face82,962 , 21CC BY-NC-SA 4.0Commercial use may be conditional
FunctionChat-BenchKorean LLM Benchmark DatasetGitHub500 + 45Apache 2.0Commercial use allowed (catalog)
KBLKorean QA DatasetHugging Face8.1MBCC BY-NC 4.0Commercial use may be conditional

Other collections