Mistral-7B-KNUT-v0.4: benchmark results
Provider: Mistral. Access: Open.
Unified ELO 1366 ± 17, rank #2131 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-EQ-Bench | 5.3 | EQ-Bench score | 51.1 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 11.37 | Exact match, flexible extract (%) | 36.4 |
| Open Korean LLM - Ko-GPQA-Diamond | 22.22 | Normalized accuracy (%) | 35 |
| Open Korean LLM - KorNAT-Harmless | 61.49 | Normalized accuracy (%) | 29.8 |
| Open Korean LLM Leaderboard | 31.27 | Average Score (%) | 26.7 |
| Open Ko-LLM Leaderboard | 31.27 | Average (%) | 26.6 |
| Open Korean LLM - KorNAT-CKA | 19.54 | Normalized accuracy (%) | 25.3 |
| Open Korean LLM - Ko-IFEval | 20.16 | Strict accuracy, prompt/instruction mean (%) | 17.9 |
| Open Korean LLM - Ko-Winogrande | 53.28 | Accuracy (%) | 12 |
| Open Korean LLM - KorNAT-Helpful | 39.25 | Normalized accuracy (%) | 4.8 |
Interactive version: theaggregate.ai/model?slug=mistral-7b-knut-v0-4 · How It Works · Data refreshed daily, snapshot 2026-09-19.