llama2_DPO_test_v1: benchmark results
Provider: Meta. Access: Open.
Unified ELO 1378 ± 17, rank #2067 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-IFEval | 28.24 | Strict accuracy, prompt/instruction mean (%) | 46.9 |
| Open Korean LLM - Ko-EQ-Bench | 0.04 | EQ-Bench score | 43.3 |
| Open Korean LLM - KorNAT-CKA | 22.97 | Normalized accuracy (%) | 41.9 |
| Open Korean LLM - Ko-GPQA-Diamond | 22.22 | Normalized accuracy (%) | 35 |
| Open Korean LLM Leaderboard | 32.26 | Average Score (%) | 32.2 |
| Open Ko-LLM Leaderboard | 32.26 | Average (%) | 32 |
| Open Korean LLM - Ko-Winogrande | 57.46 | Accuracy (%) | 31.7 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 6.07 | Exact match, flexible extract (%) | 26.6 |
| Open Korean LLM - KorNAT-Harmless | 61.28 | Normalized accuracy (%) | 26.4 |
| Open Korean LLM - KorNAT-Helpful | 43.2 | Normalized accuracy (%) | 22.1 |
Interactive version: theaggregate.ai/model?slug=llama2-dpo-test-v1 · How It Works · Data refreshed daily, snapshot 2026-09-19.