Open Ko-LLM Leaderboard: leaderboard
Korean LLM leaderboard run by Upstage on Hugging Face (2023); Average combines Ko-ARC, Ko-HellaSwag, Ko-MMLU, Ko-TruthfulQA and Ko-CommonGen v2 (Season 1) or nine tasks such as Ko-GPQA (Season 2).
Metric: Average Score (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 1157 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | gemma2-gutenberg-27B | 55.93 | #463 |
| 2 | Gemma 2 27B (IT) | 55.49 | #645 |
| 3 | Qwen2.5-14B-Gutenberg-1e-Delta | 54.8 | #473 |
| 4 | Qwen2.5-14B-Gutenberg-Instruct-Slerpeno | 54.53 | #442 |
| 5 | Qwen2.5-Lumen-14B | 54.31 | #477 |
| 6 | SuperNova-Medius | 54.26 | #479 |
| 7 | qwen2.5-reinstruct-alternate-lumen-14B | 54.07 | #449 |
| 8 | recoilme-gemma-2-9B-v0.4 | 53.97 | #553 |
| 9 | recoilme-gemma-2-9B-v0.2 | 53.44 | #629 |
| 10 | Qwen 2.5 14B Instruct | 53.23 | #647 |
| 11 | Gemma-2-Ataraxy-Gemmasutra-9B-slerp | 53.14 | #572 |
| 12 | Rombos-LLM-V2.6-Qwen-14B | 52.99 | #516 |
| 13 | recoilme-gemma-2-9B-v0.3 | 52.42 | #586 |
| 14 | romulus-mistral-nemo-12B-simpo | 52.41 | #612 |
| 15 | lambda-qwen2.5-14B-dpo-test | 52.18 | #481 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=open-ko-llm-leaderboard · How It Works · Data refreshed daily, snapshot 2026-10-20.