KOAT-5.8B: benchmark results
Provider: Other. Access: Open.
Unified ELO 1319 ± 17, rank #2393 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-GPQA-Diamond | 26.26 | Normalized accuracy (%) | 85.5 |
| Open Korean LLM - KorNAT-CKA | 21.72 | Normalized accuracy (%) | 36.3 |
| Open Korean LLM - Ko-EQ-Bench | 0 | EQ-Bench score | 24.5 |
| Open Korean LLM - Ko-IFEval | 17.19 | Strict accuracy, prompt/instruction mean (%) | 12.7 |
| Open Korean LLM Leaderboard | 29.3 | Average Score (%) | 10.9 |
| Open Ko-LLM Leaderboard | 29.3 | Average (%) | 10.7 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 0.68 | Exact match, flexible extract (%) | 10 |
| Open Korean LLM - Ko-Winogrande | 51.85 | Accuracy (%) | 7.7 |
| Open Korean LLM - KorNAT-Harmless | 59.5 | Normalized accuracy (%) | 6.8 |
| Open Korean LLM - KorNAT-Helpful | 36.2 | Normalized accuracy (%) | 1.7 |
Interactive version: theaggregate.ai/model?slug=koat-5-8b · How It Works · Data refreshed daily, snapshot 2026-09-19.