GenAI-llama2-ko-en-instruct-v4-13B: benchmark results
Provider: Other. Access: Open.
Unified ELO 1402 ± 17, rank #1896 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - KorNAT-Harmless | 64.12 | Normalized accuracy (%) | 70.8 |
| Open Korean LLM - KorNAT-Helpful | 48.55 | Normalized accuracy (%) | 69.7 |
| Open Korean LLM - Ko-Winogrande | 62.59 | Accuracy (%) | 69.2 |
| Open Korean LLM - Ko-IFEval | 30.78 | Strict accuracy, prompt/instruction mean (%) | 58 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 16.6 | Exact match, flexible extract (%) | 47.1 |
| Open Korean LLM Leaderboard | 33.36 | Average Score (%) | 41.3 |
| Open Ko-LLM Leaderboard | 33.36 | Average (%) | 41.1 |
| Open Korean LLM - KorNAT-CKA | 22 | Normalized accuracy (%) | 37.8 |
| Open Korean LLM - Ko-EQ-Bench | 0 | EQ-Bench score | 24.5 |
| Open Korean LLM - Ko-GPQA-Diamond | 18.69 | Normalized accuracy (%) | 4.1 |
Interactive version: theaggregate.ai/model?slug=genai-llama2-ko-en-instruct-v4-13b · How It Works · Data refreshed daily, snapshot 2026-09-19.