AISquare-Instruct-llama2-koen-13B-v0.9.4: benchmark results
Provider: Other. Access: Open.
Unified ELO 1390 ± 17, rank #1979 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-Winogrande | 62.83 | Accuracy (%) | 70.8 |
| Open Korean LLM - Ko-IFEval | 28.34 | Strict accuracy, prompt/instruction mean (%) | 47.4 |
| Open Korean LLM - KorNAT-CKA | 23.17 | Normalized accuracy (%) | 43.3 |
| Open Korean LLM - KorNAT-Helpful | 45.88 | Normalized accuracy (%) | 40.1 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 11.3 | Exact match, flexible extract (%) | 36.2 |
| Open Korean LLM Leaderboard | 32.74 | Average Score (%) | 35.7 |
| Open Korean LLM - KorNAT-Harmless | 61.9 | Normalized accuracy (%) | 35.6 |
| Open Ko-LLM Leaderboard | 32.74 | Average (%) | 35.5 |
| Open Korean LLM - Ko-EQ-Bench | 0 | EQ-Bench score | 24.5 |
| Open Korean LLM - Ko-GPQA-Diamond | 20.71 | Normalized accuracy (%) | 18.8 |
Interactive version: theaggregate.ai/model?slug=aisquare-instruct-llama2-koen-13b-v0-9-4 · How It Works · Data refreshed daily, snapshot 2026-09-19.