llama3-eng-ko-8: benchmark results
Provider: Meta. Access: Open.
Unified ELO 1378 ± 18, rank #2063 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-IFEval | 42.18 | Strict accuracy, prompt/instruction mean (%) | 85.7 |
| Open Korean LLM - Ko-EQ-Bench | 0.58 | EQ-Bench score | 45.8 |
| Open Korean LLM - Ko-GPQA-Diamond | 22.73 | Normalized accuracy (%) | 40.9 |
| Open Korean LLM Leaderboard | 32.51 | Average Score (%) | 34.3 |
| Open Ko-LLM Leaderboard | 32.51 | Average (%) | 34 |
| Open Korean LLM - KorNAT-Helpful | 44.67 | Normalized accuracy (%) | 31.9 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 6.07 | Exact match, flexible extract (%) | 26.6 |
| Open Korean LLM - KorNAT-CKA | 18.38 | Normalized accuracy (%) | 19.4 |
| Open Korean LLM - Ko-Winogrande | 52.41 | Accuracy (%) | 9.1 |
| Open Korean LLM - KorNAT-Harmless | 58.72 | Normalized accuracy (%) | 4.1 |
Interactive version: theaggregate.ai/model?slug=llama3-eng-ko-8 · How It Works · Data refreshed daily, snapshot 2026-09-19.