Mistral-Nemo-NT-Ko-12B-dpo: benchmark results
Provider: Mistral. Access: Open.
Unified ELO 1542 ± 17, rank #877 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-GSM8K (flexible extract) | 64.06 | Exact match, flexible extract (%) | 96.3 |
| Open Ko-LLM Leaderboard | 46.11 | Average (%) | 92.6 |
| Open Korean LLM Leaderboard | 46.11 | Average Score (%) | 92.6 |
| Open Korean LLM - Ko-GPQA-Diamond | 27.27 | Normalized accuracy (%) | 92.4 |
| Open Korean LLM - KorNAT-Helpful | 50.08 | Normalized accuracy (%) | 83.9 |
| Open Korean LLM - Ko-EQ-Bench | 32.21 | EQ-Bench score | 81.6 |
| Open Korean LLM - KorNAT-CKA | 29.84 | Normalized accuracy (%) | 81.4 |
| Open Korean LLM - Ko-IFEval | 36.61 | Strict accuracy, prompt/instruction mean (%) | 76.1 |
| Open Korean LLM - KorNAT-Harmless | 64.62 | Normalized accuracy (%) | 75.5 |
| Open Korean LLM - Ko-Winogrande | 62.51 | Accuracy (%) | 68.8 |
Interactive version: theaggregate.ai/model?slug=mistral-nemo-nt-ko-12b-dpo · How It Works · Data refreshed daily, snapshot 2026-09-19.