Nous-Hermes-2-Yi-34B: benchmark results
Nous Research's Hermes 2 fine-tune of 01.AI's Yi-34B, trained on ~1M mostly GPT-4-generated samples with ChatML formatting (December 2023). Provider: Nous Research. Released 2023-12-23. Access: Open.
Aggregate Rating 1489 ± 23, rank #812 of 1656 rated models, from 20 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - MMLU | 76.7 | Accuracy (%) (5-shot) | 97.9 |
| Open Chinese LLM - WinoGrande | 70.56 | Accuracy (%) | 94.2 |
| Open LLM Leaderboard v1 - GSM8K | 70.05 | Accuracy (%) (5-shot) | 91.4 |
| Open Chinese LLM - HellaSwag | 69.02 | Accuracy (%) | 90.2 |
| Open Chinese LLM - C-Eval Semantic | 88.73 | Accuracy (%) | 88.8 |
| Open Portuguese LLM - OAB Exams | 55.99 | Accuracy (%) | 88.8 |
| Open Chinese LLM - ARC Challenge | 59.98 | Accuracy (%) | 87.3 |
| Open Portuguese LLM - BLUEX | 65.79 | Accuracy (%) | 87.3 |
| Open LLM Leaderboard v1 - WinoGrande | 82.95 | Accuracy (%) (5-shot) | 86.4 |
| Open Chinese LLM - CMMLU | 70.04 | Accuracy (%) | 85.2 |
| Open Portuguese LLM - ENEM | 73.13 | Accuracy (%) | 84.3 |
| Open Chinese LLM - GSM8K | 57.16 | Accuracy (%) | 79.3 |
Interactive version: theaggregate.ai/model?slug=nous-hermes-2-yi-34b · How It Works · Data refreshed daily, snapshot 2026-10-20.