mistral-orpo-capybara-3k: benchmark results
Provider: Mistral. Access: Open.
Unified ELO 1458 ± 20, rank #1656 of 2928 rated models, from 11 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - HellaSwag | 85.98 | Normalized accuracy (%) (10-shot) | 77.8 |
| Open Portuguese LLM - FaQuAD NLI | 72.6 | Macro F1 (%) | 66.1 |
| Open LLM Leaderboard v1 - ARC Challenge | 63.57 | Normalized accuracy (%) (25-shot) | 63.3 |
| Open LLM Leaderboard v1 - MMLU | 62.91 | Accuracy (%) (5-shot) | 57.4 |
| Open LLM Leaderboard v1 - WinoGrande | 77.51 | Accuracy (%) (5-shot) | 56.4 |
| Open Portuguese LLM - BLUEX | 52.43 | Accuracy (%) | 52.8 |
| Open LLM Leaderboard v1 - GSM8K | 36.85 | Accuracy (%) (5-shot) | 52.2 |
| Open Portuguese LLM - OAB Exams | 40.55 | Accuracy (%) | 44.6 |
| Open Portuguese LLM - ENEM | 58.15 | Accuracy (%) | 43.3 |
| Open Portuguese LLM - ASSIN2 RTE | 86.64 | Macro F1 (%) | 38.3 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 43.83 | MC2 (%) (0-shot) | 24.9 |
Interactive version: theaggregate.ai/model?slug=mistral-orpo-capybara-3k · How It Works · Data refreshed daily, snapshot 2026-09-23.