llama-2-ko-7B-instruct: benchmark results
Provider: Meta. Access: Open.
Unified ELO 1334 ± 17, rank #2322 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-GPQA-Diamond | 23.23 | Normalized accuracy (%) | 48.1 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 5.76 | Exact match, flexible extract (%) | 25.3 |
| Open Korean LLM - Ko-Winogrande | 56.12 | Accuracy (%) | 24.7 |
| Open Korean LLM - Ko-EQ-Bench | 0 | EQ-Bench score | 24.5 |
| Open Korean LLM - KorNAT-Helpful | 43.43 | Normalized accuracy (%) | 23.4 |
| Open Korean LLM - KorNAT-CKA | 18.92 | Normalized accuracy (%) | 21.2 |
| Open Korean LLM - KorNAT-Harmless | 60.35 | Normalized accuracy (%) | 14 |
| Open Korean LLM Leaderboard | 29.71 | Average Score (%) | 13.2 |
| Open Ko-LLM Leaderboard | 29.71 | Average (%) | 13.1 |
| Open Korean LLM - Ko-IFEval | 13.75 | Strict accuracy, prompt/instruction mean (%) | 9.4 |
Interactive version: theaggregate.ai/model?slug=llama-2-ko-7b-instruct · How It Works · Data refreshed daily, snapshot 2026-09-19.