llama3-eng-ko-8B-sl3: benchmark results
Provider: Meta. Access: Open.
Unified ELO 1412 ± 17, rank #1814 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-IFEval | 37.45 | Strict accuracy, prompt/instruction mean (%) | 78.6 |
| Open Korean LLM - KorNAT-Helpful | 49.1 | Normalized accuracy (%) | 76.5 |
| Open Korean LLM - Ko-GPQA-Diamond | 24.75 | Normalized accuracy (%) | 69.5 |
| Open Korean LLM Leaderboard | 34.34 | Average Score (%) | 47.5 |
| Open Ko-LLM Leaderboard | 34.34 | Average (%) | 47.3 |
| Open Korean LLM - Ko-EQ-Bench | 0.5 | EQ-Bench score | 45.3 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 15.01 | Exact match, flexible extract (%) | 44.1 |
| Open Korean LLM - KorNAT-Harmless | 62.15 | Normalized accuracy (%) | 39.3 |
| Open Korean LLM - Ko-Winogrande | 56.99 | Accuracy (%) | 29 |
| Open Korean LLM - KorNAT-CKA | 19.41 | Normalized accuracy (%) | 24.2 |
Interactive version: theaggregate.ai/model?slug=llama3-eng-ko-8b-sl3 · How It Works · Data refreshed daily, snapshot 2026-09-19.