CapybaraHermes-2.5-Mistral-7B: benchmark results
Provider: Other. Access: Open.
Unified ELO 1475 ± 20, rank #1514 of 2928 rated models, from 11 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - HellaSwag | 85.45 | Normalized accuracy (%) (10-shot) | 74.5 |
| Open Portuguese LLM - FaQuAD NLI | 74.9 | Macro F1 (%) | 71.8 |
| Open LLM Leaderboard v1 - ARC Challenge | 65.78 | Normalized accuracy (%) (25-shot) | 71.1 |
| Open LLM Leaderboard v1 - GSM8K | 59.29 | Accuracy (%) (5-shot) | 71.1 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 56.91 | MC2 (%) (0-shot) | 69 |
| Open LLM Leaderboard v1 - WinoGrande | 78.3 | Accuracy (%) (5-shot) | 62 |
| Open LLM Leaderboard v1 - MMLU | 63.13 | Accuracy (%) (5-shot) | 58.6 |
| Open Portuguese LLM - OAB Exams | 44.33 | Accuracy (%) | 57.9 |
| Open Portuguese LLM - ASSIN2 RTE | 90.1 | Macro F1 (%) | 54.1 |
| Open Portuguese LLM - BLUEX | 52.43 | Accuracy (%) | 52.8 |
| Open Portuguese LLM - ENEM | 62.42 | Accuracy (%) | 51.6 |
Interactive version: theaggregate.ai/model?slug=capybarahermes-2-5-mistral-7b · How It Works · Data refreshed daily, snapshot 2026-09-23.