llama2_koen_13b_SFTtrain: benchmark results
Provider: Meta. Access: Open.
Unified ELO 1356 ± 17, rank #2191 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-GPQA-Diamond | 23.74 | Normalized accuracy (%) | 55.6 |
| Open Korean LLM - KorNAT-CKA | 21.37 | Normalized accuracy (%) | 34.5 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 10.31 | Exact match, flexible extract (%) | 34.1 |
| Open Korean LLM - KorNAT-Harmless | 61.45 | Normalized accuracy (%) | 28.9 |
| Open Korean LLM - Ko-Winogrande | 56.67 | Accuracy (%) | 27.4 |
| Open Korean LLM - Ko-EQ-Bench | 0 | EQ-Bench score | 24.5 |
| Open Korean LLM - KorNAT-Helpful | 43.3 | Normalized accuracy (%) | 22.9 |
| Open Korean LLM Leaderboard | 30.68 | Average Score (%) | 20.8 |
| Open Ko-LLM Leaderboard | 30.68 | Average (%) | 20.7 |
| Open Korean LLM - Ko-IFEval | 10.75 | Strict accuracy, prompt/instruction mean (%) | 8.2 |
Interactive version: theaggregate.ai/model?slug=llama2-koen-13b-sfttrain · How It Works · Data refreshed daily, snapshot 2026-09-19.