llama-2-ko-7B-pru: benchmark results
Provider: Meta. Access: Open.
Unified ELO 1254 ± 18, rank #2574 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 - Ko-EQ-Bench | 0 | EQ-Bench score | 24.5 |
| Open Korean LLM - KorNAT-Helpful | 42.6 | Normalized accuracy (%) | 17.6 |
| Open Korean LLM - KorNAT-Harmless | 60.37 | Normalized accuracy (%) | 14.1 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 1.52 | Exact match, flexible extract (%) | 13.9 |
| Open Korean LLM - Ko-Winogrande | 51.14 | Accuracy (%) | 5.5 |
| Open Korean LLM - KorNAT-CKA | 10.24 | Normalized accuracy (%) | 1.9 |
| Open Korean LLM Leaderboard | 26.52 | Average Score (%) | 1.3 |
| Open Ko-LLM Leaderboard | 26.52 | Average (%) | 1.2 |
| Open Korean LLM - Ko-IFEval | 0.15 | Strict accuracy, prompt/instruction mean (%) | 0.8 |
Interactive version: theaggregate.ai/model?slug=llama-2-ko-7b-pru · How It Works · Data refreshed daily, snapshot 2026-09-19.