Llama 2 13B Dpo V2: benchmark results
Provider: Meta. Access: Open.
Unified ELO 1407 ± 17, rank #1847 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - KorNAT-Harmless | 65.08 | Normalized accuracy (%) | 78.9 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 21.08 | Exact match, flexible extract (%) | 53.4 |
| Open Korean LLM - Ko-IFEval | 29.02 | Strict accuracy, prompt/instruction mean (%) | 50.2 |
| Open Korean LLM Leaderboard | 34.03 | Average Score (%) | 46 |
| Open Ko-LLM Leaderboard | 34.03 | Average (%) | 45.9 |
| Open Korean LLM - Ko-Winogrande | 57.85 | Accuracy (%) | 34.9 |
| Open Korean LLM - Ko-GPQA-Diamond | 21.72 | Normalized accuracy (%) | 29.4 |
| Open Korean LLM - KorNAT-CKA | 20.37 | Normalized accuracy (%) | 29.4 |
| Open Korean LLM - KorNAT-Helpful | 43.77 | Normalized accuracy (%) | 25.3 |
| Open Korean LLM - Ko-EQ-Bench | 0 | EQ-Bench score | 24.5 |
Interactive version: theaggregate.ai/model?slug=llama-2-13b-dpo-v2 · How It Works · Data refreshed daily, snapshot 2026-09-19.