polyglot-ko-3.8B-chat: benchmark results
Provider: Other. Access: Open.
Unified ELO 1356 ± 18, rank #2190 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-GPQA-Diamond | 27.27 | Normalized accuracy (%) | 92.4 |
| Open Korean LLM - KorNAT-CKA | 22.09 | Normalized accuracy (%) | 38.1 |
| Open Korean LLM - Ko-IFEval | 25.17 | Strict accuracy, prompt/instruction mean (%) | 33.4 |
| Open Korean LLM - Ko-EQ-Bench | 0 | EQ-Bench score | 24.5 |
| Open Korean LLM Leaderboard | 31.04 | Average Score (%) | 24.3 |
| Open Ko-LLM Leaderboard | 31.04 | Average (%) | 24 |
| Open Korean LLM - Ko-Winogrande | 52.88 | Accuracy (%) | 11.1 |
| Open Korean LLM - KorNAT-Harmless | 59.89 | Normalized accuracy (%) | 10.1 |
| Open Korean LLM - KorNAT-Helpful | 41.08 | Normalized accuracy (%) | 10.1 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 0 | Exact match, flexible extract (%) | 2 |
Interactive version: theaggregate.ai/model?slug=polyglot-ko-3-8b-chat · How It Works · Data refreshed daily, snapshot 2026-09-19.