llama3-open-ko-8B-Instruct-shimshimi-500-ver2: benchmark results
Provider: Meta. Access: Open.
Unified ELO 1413 ± 17, rank #1807 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - KorNAT-CKA | 32.91 | Normalized accuracy (%) | 89.9 |
| Open Korean LLM - Ko-GPQA-Diamond | 25.25 | Normalized accuracy (%) | 75.9 |
| Open Korean LLM - Ko-IFEval | 30.46 | Strict accuracy, prompt/instruction mean (%) | 56.3 |
| Open Korean LLM - Ko-EQ-Bench | 4.22 | EQ-Bench score | 50.4 |
| Open Korean LLM - KorNAT-Helpful | 46.9 | Normalized accuracy (%) | 50.3 |
| Open Korean LLM Leaderboard | 34.1 | Average Score (%) | 46.4 |
| Open Ko-LLM Leaderboard | 34.1 | Average (%) | 46.2 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 13.19 | Exact match, flexible extract (%) | 40.4 |
| Open Korean LLM - KorNAT-Harmless | 61.81 | Normalized accuracy (%) | 34.3 |
| Open Korean LLM - Ko-Winogrande | 57.14 | Accuracy (%) | 29.8 |
Interactive version: theaggregate.ai/model?slug=llama3-open-ko-8b-instruct-shimshimi-500-ver2 · How It Works · Data refreshed daily, snapshot 2026-09-19.