AISquare-Instruct-llama2-koen-13B-v0.9.12: benchmark results
Provider: Other. Access: Open.
Unified ELO 1409 ± 17, rank #1833 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-Winogrande | 64.4 | Accuracy (%) | 78.4 |
| Open Korean LLM - KorNAT-CKA | 24.67 | Normalized accuracy (%) | 52.7 |
| Open Korean LLM Leaderboard | 34.06 | Average Score (%) | 46.1 |
| Open Ko-LLM Leaderboard | 34.06 | Average (%) | 45.9 |
| Open Korean LLM - Ko-IFEval | 27.98 | Strict accuracy, prompt/instruction mean (%) | 45.8 |
| Open Korean LLM - KorNAT-Helpful | 45.77 | Normalized accuracy (%) | 39.2 |
| Open Korean LLM - KorNAT-Harmless | 61.82 | Normalized accuracy (%) | 34.5 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 10.46 | Exact match, flexible extract (%) | 34.2 |
| Open Korean LLM - Ko-GPQA-Diamond | 21.72 | Normalized accuracy (%) | 29.4 |
| Open Korean LLM - Ko-EQ-Bench | 0 | EQ-Bench score | 24.5 |
Interactive version: theaggregate.ai/model?slug=aisquare-instruct-llama2-koen-13b-v0-9-12 · How It Works · Data refreshed daily, snapshot 2026-09-19.