Llama-3-Open-Ko-8B-Instruct-seongi-ver0710: benchmark results
Provider: Meta. Access: Open.
Unified ELO 1401 ± 18, rank #1899 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-GPQA-Diamond | 28.79 | Normalized accuracy (%) | 98.1 |
| Open Korean LLM - KorNAT-Harmless | 65.19 | Normalized accuracy (%) | 79.4 |
| Open Korean LLM - KorNAT-Helpful | 48.58 | Normalized accuracy (%) | 69.9 |
| Open Korean LLM - KorNAT-CKA | 24.52 | Normalized accuracy (%) | 51.8 |
| Open Korean LLM - Ko-EQ-Bench | 3.85 | EQ-Bench score | 50.1 |
| Open Korean LLM Leaderboard | 33.6 | Average Score (%) | 43 |
| Open Ko-LLM Leaderboard | 33.6 | Average (%) | 42.7 |
| Open Korean LLM - Ko-IFEval | 25.84 | Strict accuracy, prompt/instruction mean (%) | 36.9 |
| Open Korean LLM - Ko-Winogrande | 55.88 | Accuracy (%) | 22.9 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 1.74 | Exact match, flexible extract (%) | 15.6 |
Interactive version: theaggregate.ai/model?slug=llama-3-open-ko-8b-instruct-seongi-ver0710 · How It Works · Data refreshed daily, snapshot 2026-09-19.