Hermes-2-Pro-Mistral-7B: benchmark results
Nous Research's Hermes 2 Pro Mistral 7B, retrained from OpenHermes 2.5 data plus a new in-house function-calling and JSON-mode dataset, using a dedicated ChatML role for structured tool calls. Provider: Nous Research. Released 2024-03-14. Access: Open.
Aggregate Rating 1398 ± 21, rank #1230 of 1656 rated models, from 31 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard - MuSR | 43.76 | Score | 76.5 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 58.99 | MC2 (%) (0-shot) | 72.9 |
| Open LLM Leaderboard v1 - GSM8K | 60.42 | Accuracy (%) (5-shot) | 72.5 |
| Open LLM Leaderboard - IFEval | 56.68 | Score | 68.5 |
| Open LLM Leaderboard v1 - ARC Challenge | 64.16 | Normalized accuracy (%) (25-shot) | 65.7 |
| Open Medical LLM - PubMedQA | 74.6 | Accuracy (%) | 62.5 |
| Open LLM Leaderboard v1 - MMLU | 62.21 | Accuracy (%) (5-shot) | 54.4 |
| Open LLM Leaderboard v1 - HellaSwag | 82.73 | Normalized accuracy (%) (10-shot) | 53.4 |
| YALL Nous Leaderboard | 52.55 | Average | 49.4 |
| Open LLM Leaderboard - BBH | 49.95 | Score | 48.2 |
| Open Medical LLM - MMLU College Medicine | 61.85 | Accuracy (%) | 47.7 |
| Open CoT - LSAT Logical Reasoning | 9.41 | CoT Gain (%) | 45 |
Interactive version: theaggregate.ai/model?slug=hermes-2-pro-mistral-7b · How It Works · Data refreshed daily, snapshot 2026-10-20.