Hermes 2 models
4 Hermes 2 models from Nous Research on Hugging Face, from 6.7B to 46.7B parameters, published by Nous Research in BF16, with contexts from 4K to 8K tokens. The smallest official model, Nous-Hermes-llama-2-7b, needs about 15.1 GB of VRAM at its published precision; the cheapest live fit is RTX A4000 at $0.167/hr.
Pick a size
One row per official Hermes 2 size: the VRAM it needs at each precision, the cheapest GPU that holds it today, and the KV cache for a 32K-token context.
| Model | Parameters | Native VRAM | FP8 VRAM | INT4 VRAM | Live GPU fit (native) | Est. $/hr | KV cache at 32K |
|---|---|---|---|---|---|---|---|
| Nous-Hermes-llama-2-7b | 6.7B | 15.1 GB | 7.5 GB | 3.8 GB | RTX A4000 | $0.167/hr | 16.0 GB |
| Hermes-2-Theta-Llama-3-8B | 8.0B | 17.9 GB | 9.0 GB | 4.5 GB | RTX A5000 | $0.176/hr | 4.0 GB |
VRAM is weights times a flat 1.2 overhead; the KV cache is a separate per-model figure at 16-bit, shown where the architecture is published. See the methodology.
Official models (2)
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| Nous-Hermes-llama-2-7b | 6.7B | BF16 | 15.1 GB | RTX A4000 | $0.167/hr |
| Hermes-2-Theta-Llama-3-8B | 8.0B | BF16 | 17.9 GB | RTX A5000 | $0.176/hr |
Fine-tunes and community models (2)
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| Nous-Hermes-2-Mistral-7B-DPO | 7.2B | BF16 | 16.2 GB | RTX A5000 | $0.176/hr |
| Nous-Hermes-2-Mixtral-8x7B-DPO | 46.7B | BF16 | 104 GB | AMD MI300X | $2.63/hr |
VRAM is for the precision the model is published in: the weights times a flat 1.2 overhead, with the KV cache not included. See the methodology. The fit shown is the lowest-priced single GPU type that holds the model at that precision, or the lowest-priced multi-GPU set (up to 8) when none does.
More from Nous Research
- Hermes 3 (3 models)