llama3-eng-ko-8B-sl2: benchmark results
Provider: Meta. Access: Open.
Unified ELO 1430 ± 17, rank #1669 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-IFEval | 40.64 | Strict accuracy, prompt/instruction mean (%) | 83.2 |
| Open Korean LLM - KorNAT-Helpful | 48.27 | Normalized accuracy (%) | 66.9 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 26.54 | Exact match, flexible extract (%) | 60.5 |
| Open Korean LLM Leaderboard | 36.4 | Average Score (%) | 57.1 |
| Open Ko-LLM Leaderboard | 36.4 | Average (%) | 57 |
| Open Korean LLM - Ko-EQ-Bench | 7.01 | EQ-Bench score | 53.1 |
| Open Korean LLM - Ko-GPQA-Diamond | 22.73 | Normalized accuracy (%) | 40.9 |
| Open Korean LLM - KorNAT-Harmless | 61.68 | Normalized accuracy (%) | 32.5 |
| Open Korean LLM - Ko-Winogrande | 55.64 | Accuracy (%) | 21.5 |
| Open Korean LLM - KorNAT-CKA | 16.69 | Normalized accuracy (%) | 13 |
Interactive version: theaggregate.ai/model?slug=llama3-eng-ko-8b-sl2 · How It Works · Data refreshed daily, snapshot 2026-09-19.