Llama 2 13B Dpo V3: benchmark results
Provider: Meta. Access: Open.
Unified ELO 1433 ± 17, rank #1644 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-Winogrande | 65.82 | Accuracy (%) | 86.5 |
| Open Korean LLM - KorNAT-Helpful | 50.55 | Normalized accuracy (%) | 86.3 |
| Open Korean LLM - KorNAT-Harmless | 66.45 | Normalized accuracy (%) | 86.1 |
| Open Korean LLM - KorNAT-CKA | 26.22 | Normalized accuracy (%) | 63.1 |
| Open Korean LLM Leaderboard | 36.08 | Average Score (%) | 55.7 |
| Open Ko-LLM Leaderboard | 36.08 | Average (%) | 55.6 |
| Open Korean LLM - Ko-IFEval | 30.19 | Strict accuracy, prompt/instruction mean (%) | 54.8 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 17.06 | Exact match, flexible extract (%) | 47.8 |
| Open Korean LLM - Ko-EQ-Bench | 0 | EQ-Bench score | 24.5 |
| Open Korean LLM - Ko-GPQA-Diamond | 18.18 | Normalized accuracy (%) | 2.3 |
Interactive version: theaggregate.ai/model?slug=llama-2-13b-dpo-v3 · How It Works · Data refreshed daily, snapshot 2026-09-19.