ko-ref-llama2-7B: benchmark results
Provider: Other. Access: Open.
Unified ELO 1320 ± 17, rank #2388 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-GPQA-Diamond | 23.74 | Normalized accuracy (%) | 55.6 |
| Open Korean LLM - Ko-Winogrande | 59.75 | Accuracy (%) | 48.8 |
| Open Korean LLM - Ko-EQ-Bench | 0 | EQ-Bench score | 24.5 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 3.41 | Exact match, flexible extract (%) | 21.4 |
| Open Korean LLM - KorNAT-Harmless | 60.94 | Normalized accuracy (%) | 21.1 |
| Open Korean LLM - KorNAT-Helpful | 42.48 | Normalized accuracy (%) | 16.8 |
| Open Korean LLM Leaderboard | 29.12 | Average Score (%) | 9.9 |
| Open Ko-LLM Leaderboard | 29.12 | Average (%) | 9.8 |
| Open Korean LLM - KorNAT-CKA | 15.05 | Normalized accuracy (%) | 7.7 |
| Open Korean LLM - Ko-IFEval | 9.09 | Strict accuracy, prompt/instruction mean (%) | 7.5 |
Interactive version: theaggregate.ai/model?slug=ko-ref-llama2-7b · How It Works · Data refreshed daily, snapshot 2026-09-19.