Mistral-7B-Instruct-SimPO: benchmark results
Provider: Princeton NLP. Released 2024-05-24. Access: Open.
Aggregate Rating 1376 ± 19, rank #1310 of 1661 rated models, from 21 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - KorNAT-Harmless | 66.65 | Normalized accuracy (%) | 86.5 |
| Open Korean LLM - KorNAT-Helpful | 49.3 | Normalized accuracy (%) | 78.3 |
| Open Korean LLM - Ko-GPQA-Diamond | 24.75 | Normalized accuracy (%) | 69.5 |
| Open Korean LLM - Ko-EQ-Bench | 17.53 | EQ-Bench score | 63.4 |
| Open Korean LLM - Ko-IFEval | 32.01 | Strict accuracy, prompt/instruction mean (%) | 61.3 |
| Open Portuguese LLM - BLUEX | 53.41 | Accuracy (%) | 55.9 |
| Open LLM Leaderboard - IFEval | 46.87 | Score | 53.5 |
| Open Portuguese LLM - FaQuAD NLI | 66.55 | Macro F1 (%) | 52.7 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 20.62 | Exact match, flexible extract (%) | 52.6 |
| Open Korean LLM Leaderboard | 35.24 | Average Score (%) | 50.8 |
| Open Ko-LLM Leaderboard | 35.24 | Average (%) | 50.7 |
| Open LLM Leaderboard - MuSR | 40.98 | Score | 50.2 |
Interactive version: theaggregate.ai/model?slug=mistral-7b-instruct-simpo · How It Works · Data refreshed daily, snapshot 2026-10-23.