Llama 2 13B .kor.v1: benchmark results
Provider: Meta. Access: Open.
Unified ELO 1360 ± 17, rank #2174 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-Winogrande | 58.33 | Accuracy (%) | 37.8 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 11.52 | Exact match, flexible extract (%) | 36.8 |
| Open Korean LLM - Ko-IFEval | 24.74 | Strict accuracy, prompt/instruction mean (%) | 31 |
| Open Korean LLM - Ko-EQ-Bench | 0 | EQ-Bench score | 24.5 |
| Open Korean LLM Leaderboard | 31.04 | Average Score (%) | 24.3 |
| Open Korean LLM - KorNAT-Helpful | 43.55 | Normalized accuracy (%) | 24.2 |
| Open Ko-LLM Leaderboard | 31.04 | Average (%) | 24.1 |
| Open Korean LLM - KorNAT-Harmless | 60.28 | Normalized accuracy (%) | 13.2 |
| Open Korean LLM - Ko-GPQA-Diamond | 19.7 | Normalized accuracy (%) | 9.6 |
| Open Korean LLM - KorNAT-CKA | 13.28 | Normalized accuracy (%) | 4.6 |
Interactive version: theaggregate.ai/model?slug=llama-2-13b-kor-v1 · How It Works · Data refreshed daily, snapshot 2026-09-19.