llama3-eng-ko-8B-sl4: benchmark results
Provider: Meta. Access: Open.
Unified ELO 1429 ± 17, rank #1679 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - KorNAT-Helpful | 49.28 | Normalized accuracy (%) | 78.1 |
| Open Korean LLM - Ko-IFEval | 37.3 | Strict accuracy, prompt/instruction mean (%) | 78 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 29.8 | Exact match, flexible extract (%) | 64.3 |
| Open Korean LLM Leaderboard | 36.25 | Average Score (%) | 56.6 |
| Open Ko-LLM Leaderboard | 36.25 | Average (%) | 56.5 |
| Open Korean LLM - Ko-EQ-Bench | 8.91 | EQ-Bench score | 55.1 |
| Open Korean LLM - KorNAT-Harmless | 62.74 | Normalized accuracy (%) | 49.6 |
| Open Korean LLM - KorNAT-CKA | 18.11 | Normalized accuracy (%) | 17.9 |
| Open Korean LLM - Ko-Winogrande | 54.46 | Accuracy (%) | 16.3 |
| Open Korean LLM - Ko-GPQA-Diamond | 18.69 | Normalized accuracy (%) | 4.1 |
Interactive version: theaggregate.ai/model?slug=llama3-eng-ko-8b-sl4 · How It Works · Data refreshed daily, snapshot 2026-09-19.