Llama 3 8B Dpo Sft V1: benchmark results
Provider: Meta. Access: Open.
Unified ELO 1473 ± 17, rank #1352 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-GPQA-Diamond | 26.77 | Normalized accuracy (%) | 88.9 |
| Open Korean LLM - Ko-IFEval | 37.35 | Strict accuracy, prompt/instruction mean (%) | 78.1 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 41.02 | Exact match, flexible extract (%) | 75.4 |
| Open Korean LLM - Ko-EQ-Bench | 26.74 | EQ-Bench score | 73.5 |
| Open Korean LLM Leaderboard | 39.54 | Average Score (%) | 72.7 |
| Open Ko-LLM Leaderboard | 39.54 | Average (%) | 72.4 |
| Open Korean LLM - KorNAT-CKA | 25.03 | Normalized accuracy (%) | 54.7 |
| Open Korean LLM - Ko-Winogrande | 60.06 | Accuracy (%) | 51.1 |
| Open Korean LLM - KorNAT-Helpful | 45.98 | Normalized accuracy (%) | 41 |
| Open Korean LLM - KorNAT-Harmless | 61.3 | Normalized accuracy (%) | 26.9 |
Interactive version: theaggregate.ai/model?slug=llama-3-8b-dpo-sft-v1 · How It Works · Data refreshed daily, snapshot 2026-09-19.