mistral-7B-orpo-alignment-handbook: benchmark results
Provider: Mistral. Access: Open.
Unified ELO 1462 ± 20, rank #1616 of 2928 rated models, from 11 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - HellaSwag | 83.96 | Normalized accuracy (%) (10-shot) | 62.7 |
| Open LLM Leaderboard v1 - MMLU | 63.49 | Accuracy (%) (5-shot) | 61.1 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 53.87 | MC2 (%) (0-shot) | 61 |
| Open LLM Leaderboard v1 - WinoGrande | 77.82 | Accuracy (%) (5-shot) | 58.7 |
| Open LLM Leaderboard v1 - ARC Challenge | 62.46 | Normalized accuracy (%) (25-shot) | 58.1 |
| Open LLM Leaderboard v1 - GSM8K | 41.32 | Accuracy (%) (5-shot) | 56.4 |
| Open Portuguese LLM - OAB Exams | 41.32 | Accuracy (%) | 46.6 |
| Open Portuguese LLM - ENEM | 59.55 | Accuracy (%) | 45.8 |
| Open Portuguese LLM - BLUEX | 49.65 | Accuracy (%) | 45.6 |
| Open Portuguese LLM - FaQuAD NLI | 45.45 | Macro F1 (%) | 28.4 |
| Open Portuguese LLM - ASSIN2 RTE | 76.54 | Macro F1 (%) | 28.2 |
Interactive version: theaggregate.ai/model?slug=mistral-7b-orpo-alignment-handbook · How It Works · Data refreshed daily, snapshot 2026-09-23.