Llama-3-Open-Ko-8B-Instruct-preview: benchmark results
Provider: Meta. Access: Open.
Unified ELO 1419 ± 17, rank #1755 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - KorNAT-CKA | 30.34 | Normalized accuracy (%) | 82.7 |
| Open Korean LLM - KorNAT-Harmless | 64.86 | Normalized accuracy (%) | 77.3 |
| Open Korean LLM - Ko-IFEval | 34.44 | Strict accuracy, prompt/instruction mean (%) | 70.9 |
| Open Korean LLM - Ko-GPQA-Diamond | 24.24 | Normalized accuracy (%) | 62.4 |
| Open Korean LLM - Ko-EQ-Bench | 8.09 | EQ-Bench score | 54.5 |
| Open Korean LLM - KorNAT-Helpful | 46.83 | Normalized accuracy (%) | 49.7 |
| Open Korean LLM Leaderboard | 34.74 | Average Score (%) | 48.7 |
| Open Ko-LLM Leaderboard | 34.74 | Average (%) | 48.6 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 11.83 | Exact match, flexible extract (%) | 37.6 |
| Open Korean LLM - Ko-Winogrande | 55.96 | Accuracy (%) | 23.7 |
Interactive version: theaggregate.ai/model?slug=llama-3-open-ko-8b-instruct-preview · How It Works · Data refreshed daily, snapshot 2026-09-19.