Nous-Hermes-13B-SuperHOT-8K-fp16: benchmark results
TheBloke's Hugging Face publication of NousResearch's Nous-Hermes-13B, whose original release is NousResearch/Nous-Hermes-13b. Conversion publication is credited separately from the original model creator. The card describes a SuperHOT merge with extended-context inference requirements; its extension is not treated as the original model’s native context. Provider: TheBloke. Access: Open.
Aggregate Rating 1308 ± 16, rank #8594 of 10916 rated models, from 68 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - Machine Learning | 41.96 | Accuracy (%) | 84.6 |
| MMLU-by-task - TruthfulQA MC2 | 51.19 | Accuracy (%) | 81.6 |
| MMLU-by-task - TruthfulQA MC1 | 35.13 | Accuracy (%) | 81.1 |
| MMLU-by-task - Formal Logic | 36.51 | Accuracy (%) | 76.5 |
| MMLU-by-task - College Computer Science | 46 | Accuracy (%) | 75.6 |
| MMLU-by-task - HellaSwag | 62.3 | Accuracy (%) | 75.5 |
| MMLU-by-task - Sociology | 71.14 | Accuracy (%) | 69 |
| MMLU-by-task - US Foreign Policy | 78 | Accuracy (%) | 67.9 |
| MMLU-by-task - Human Sexuality | 60.31 | Accuracy (%) | 65.8 |
| MMLU-by-task - Medical Genetics | 56 | Accuracy (%) | 65.3 |
| MMLU-by-task - ARC Challenge | 54.01 | Accuracy (%) | 65.1 |
| MMLU-by-task - Computer Security | 65 | Accuracy (%) | 65 |
Interactive version: theaggregate.ai/model?slug=nous-hermes-13b-superhot-8k-fp16 · How It Works · Data refreshed daily, snapshot 2026-11-08.