Llama 3 8B Dpo V1: benchmark results
Provider: Meta. Access: Open.
Unified ELO 1461 ± 18, rank #1438 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-GPQA-Diamond | 27.78 | Normalized accuracy (%) | 95 |
| Open Korean LLM - KorNAT-CKA | 33.46 | Normalized accuracy (%) | 92 |
| Open Ko-LLM Leaderboard | 39.07 | Average (%) | 70 |
| Open Korean LLM Leaderboard | 39.07 | Average Score (%) | 69.9 |
| Open Korean LLM - Ko-Winogrande | 62.19 | Accuracy (%) | 67.2 |
| Open Korean LLM - Ko-EQ-Bench | 13.89 | EQ-Bench score | 59.7 |
| Open Korean LLM - KorNAT-Harmless | 63.14 | Normalized accuracy (%) | 57.3 |
| Open Korean LLM - Ko-IFEval | 27.25 | Strict accuracy, prompt/instruction mean (%) | 42.1 |
| Open Korean LLM - KorNAT-Helpful | 44.88 | Normalized accuracy (%) | 33.1 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 1.67 | Exact match, flexible extract (%) | 15 |
Interactive version: theaggregate.ai/model?slug=llama-3-8b-dpo-v1 · How It Works · Data refreshed daily, snapshot 2026-09-19.