Llama 3 8B Inst Unsloth Ko Merged 1: benchmark results
Provider: Meta. Access: Open.
Unified ELO 1424 ± 17, rank #1719 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-IFEval | 37.41 | Strict accuracy, prompt/instruction mean (%) | 78.4 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 32.6 | Exact match, flexible extract (%) | 66.4 |
| Open Korean LLM - Ko-EQ-Bench | 18.15 | EQ-Bench score | 64.7 |
| Open Korean LLM Leaderboard | 35.22 | Average Score (%) | 50.5 |
| Open Ko-LLM Leaderboard | 35.22 | Average (%) | 50.3 |
| Open Korean LLM - KorNAT-CKA | 23.74 | Normalized accuracy (%) | 46.5 |
| Open Korean LLM - KorNAT-Helpful | 44.32 | Normalized accuracy (%) | 28.8 |
| Open Korean LLM - Ko-Winogrande | 56.51 | Accuracy (%) | 26.8 |
| Open Korean LLM - Ko-GPQA-Diamond | 21.21 | Normalized accuracy (%) | 24 |
| Open Korean LLM - KorNAT-Harmless | 60.7 | Normalized accuracy (%) | 17.8 |
Interactive version: theaggregate.ai/model?slug=llama-3-8b-inst-unsloth-ko-merged-1 · How It Works · Data refreshed daily, snapshot 2026-09-19.