ko-ref-llama2-13B: benchmark results
Provider: Other. Access: Open.
Aggregate Rating 1247 ± 34, rank #9279 of 10916 rated models, from 15 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-Winogrande | 63.77 | Accuracy (%) | 75.7 |
| Open Korean LLM - Ko-GPQA-Diamond | 23.23 | Normalized accuracy (%) | 48.1 |
| Open Korean LLM - KorNAT-Harmless | 61.87 | Normalized accuracy (%) | 35.1 |
| Open Korean LLM - Ko-EQ-Bench | 0 | EQ-Bench score | 24.5 |
| Open LLM Leaderboard v1 - HellaSwag | 73.56 | Normalized accuracy (%) (10-shot) | 23.8 |
| Open LLM Leaderboard v1 - WinoGrande | 69.14 | Accuracy (%) (5-shot) | 23.2 |
| Open LLM Leaderboard v1 - ARC Challenge | 48.38 | Normalized accuracy (%) (25-shot) | 23.1 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 4.32 | Exact match, flexible extract (%) | 22.8 |
| Open LLM Leaderboard v1 - MMLU | 34.83 | Accuracy (%) (5-shot) | 18.2 |
| Open Korean LLM - KorNAT-CKA | 16.74 | Normalized accuracy (%) | 13.3 |
| Open Korean LLM - KorNAT-Helpful | 41.6 | Normalized accuracy (%) | 13 |
| Open Korean LLM Leaderboard | 29.62 | Average Score (%) | 12.5 |
Interactive version: theaggregate.ai/model?slug=ko-ref-llama2-13b · How It Works · Data refreshed daily, snapshot 2026-11-08.