llama3-koen-sft-dpo-v1: benchmark results
Provider: Meta. Access: Open.
Unified ELO 1470 ± 17, rank #1379 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - KorNAT-CKA | 34.07 | Normalized accuracy (%) | 93.4 |
| Open Korean LLM - Ko-EQ-Bench | 30.75 | EQ-Bench score | 79.8 |
| Open Korean LLM - Ko-Winogrande | 63.3 | Accuracy (%) | 73.4 |
| Open Korean LLM - Ko-GPQA-Diamond | 24.75 | Normalized accuracy (%) | 69.5 |
| Open Ko-LLM Leaderboard | 38.98 | Average (%) | 69.3 |
| Open Korean LLM Leaderboard | 38.98 | Average Score (%) | 69.3 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 28.43 | Exact match, flexible extract (%) | 62.8 |
| Open Korean LLM - KorNAT-Harmless | 62.79 | Normalized accuracy (%) | 50.3 |
| Open Korean LLM - Ko-IFEval | 27.54 | Strict accuracy, prompt/instruction mean (%) | 43.6 |
| Open Korean LLM - KorNAT-Helpful | 45.92 | Normalized accuracy (%) | 40.7 |
Interactive version: theaggregate.ai/model?slug=llama3-koen-sft-dpo-v1 · How It Works · Data refreshed daily, snapshot 2026-09-19.