Mistral-7B-KNUT-v0.3: benchmark results
Provider: Mistral. Access: Open.
Unified ELO 1351 ± 17, rank #2229 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-EQ-Bench | 3.99 | EQ-Bench score | 50.2 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 12.21 | Exact match, flexible extract (%) | 38.7 |
| Open Korean LLM - Ko-GPQA-Diamond | 22.22 | Normalized accuracy (%) | 35 |
| Open Korean LLM - KorNAT-Harmless | 61.25 | Normalized accuracy (%) | 26 |
| Open Korean LLM Leaderboard | 30.6 | Average Score (%) | 20.3 |
| Open Ko-LLM Leaderboard | 30.6 | Average (%) | 20.2 |
| Open Korean LLM - Ko-IFEval | 20.2 | Strict accuracy, prompt/instruction mean (%) | 18.1 |
| Open Korean LLM - KorNAT-CKA | 15.16 | Normalized accuracy (%) | 8.2 |
| Open Korean LLM - Ko-Winogrande | 51.22 | Accuracy (%) | 5.8 |
| Open Korean LLM - KorNAT-Helpful | 39.83 | Normalized accuracy (%) | 5.8 |
Interactive version: theaggregate.ai/model?slug=mistral-7b-knut-v0-3 · How It Works · Data refreshed daily, snapshot 2026-09-19.