Mistral-7B-Base-SFT-SimPO: benchmark results
Princeton NLP's Mistral-7B-Base-SFT-SimPO, a Mistral 7B base model with SFT and SimPO alignment, from Princeton's SimPO preference-optimization paper. Provider: Princeton NLP. Released 2024-05-17. Access: Open.
Aggregate Rating 1373 ± 20, rank #1319 of 1654 rated models, from 20 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-GPQA-Diamond | 26.77 | Normalized accuracy (%) | 88.9 |
| Open Korean LLM - Ko-IFEval | 34.22 | Strict accuracy, prompt/instruction mean (%) | 70.2 |
| Open Korean LLM - Ko-EQ-Bench | 17.58 | EQ-Bench score | 63.4 |
| Open Korean LLM - KorNAT-Harmless | 63.42 | Normalized accuracy (%) | 62.3 |
| Open LLM Leaderboard - IFEval | 47.01 | Score | 53.9 |
| Open Korean LLM Leaderboard | 35.59 | Average Score (%) | 52 |
| Open Korean LLM - KorNAT-Helpful | 46.65 | Normalized accuracy (%) | 48.1 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 16.98 | Exact match, flexible extract (%) | 47.6 |
| Open Portuguese LLM - ASSIN2 RTE | 88.75 | Macro F1 (%) | 46.4 |
| Open Portuguese LLM - FaQuAD NLI | 61.09 | Macro F1 (%) | 45.2 |
| Open Portuguese LLM - OAB Exams | 40.68 | Accuracy (%) | 44.8 |
| Open Portuguese LLM - BLUEX | 48.68 | Accuracy (%) | 43 |
Interactive version: theaggregate.ai/model?slug=mistral-7b-base-sft-simpo · How It Works · Data refreshed daily, snapshot 2026-10-24.