llama3-open-ko-8B-shimshimi: benchmark results
Provider: Meta. Access: Open.
Unified ELO 1410 ± 17, rank #1825 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - KorNAT-CKA | 33.11 | Normalized accuracy (%) | 90.5 |
| Open Korean LLM - Ko-GPQA-Diamond | 25.25 | Normalized accuracy (%) | 75.9 |
| Open Korean LLM - KorNAT-Helpful | 47.1 | Normalized accuracy (%) | 53.1 |
| Open Korean LLM - Ko-EQ-Bench | 5.27 | EQ-Bench score | 51 |
| Open Korean LLM Leaderboard | 33.82 | Average Score (%) | 44.7 |
| Open Ko-LLM Leaderboard | 33.82 | Average (%) | 44.4 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 12.89 | Exact match, flexible extract (%) | 39.8 |
| Open Korean LLM - Ko-IFEval | 25.87 | Strict accuracy, prompt/instruction mean (%) | 37 |
| Open Korean LLM - Ko-Winogrande | 56.91 | Accuracy (%) | 28.5 |
| Open Korean LLM - KorNAT-Harmless | 61.4 | Normalized accuracy (%) | 28 |
Interactive version: theaggregate.ai/model?slug=llama3-open-ko-8b-shimshimi · How It Works · Data refreshed daily, snapshot 2026-09-19.