Hermes Models — Hardware Requirements
4 Hermes models from Nous Research and the community, from the smallest that runs in 5.1 GB of VRAM up to 405.9B parameters. Every row links to full quantization tables and GPU compatibility.
All Hermes Models by Size
| Model | Params | Runs from | Context | Publisher | Quant downloads |
|---|---|---|---|---|---|
| Carnice V1 9B Hermes Agent Stage2 Merged | 9.0B | 4.4 GB | 262K | ||
| Hermes 4 14B | 14.8B | 5.1 GB | 41K | ||
| Hermes 4.3 36B | 36.2B | 10.5 GB | 524K | ||
| Hermes 4 70B | 70.6B | 31.0 GB | 131K | ||
| Hermes 4 405B | 405.9B | 173.8 GB | 131K |
Frequently Asked Questions
- How much VRAM do I need to run a Hermes model?
- The smallest Hermes model, Carnice V1 9B Hermes Agent Stage2 Merged, runs from 4.4 GB of VRAM at an aggressive quantization. Larger family members need proportionally more — see the table above for every model.
- Which Hermes models can I run on a 16 GB GPU?
- 3 of 5 Hermes models fit in 16 GB of VRAM at some quantization, including Hermes 4 14B, Hermes 4.3 36B, Carnice V1 9B Hermes Agent Stage2 Merged.
- What is the most popular Hermes model to run locally?
- Hermes 4 14B is the most downloaded Hermes model in local-friendly quantized formats. It runs from 5.1 GB of VRAM.