4-bit Open on Hugging Face
Parameters
3.2B
Weights (est.)
≈ 1.8 GB
Mac memory (est.)
8 GB+
Size is estimated from the parameter count and quantisation. Memory allows a fifth extra for context plus 2 GB for the system; long contexts need more.
Run it
With mlx-lm on an Apple silicon Mac:
pip install mlx-lmmlx_lm.generate --model mlx-community/Hermes-3-Llama-3.2-3B-4bit --prompt "Hello"
Details
- Base model
- NousResearch/Hermes-3-Llama-3.2-3B (quantized)
- Vendor
- NousResearch
- Task
- Text generation
- Quantisation
- 4-bit
- Licence
- llama3
- Downloads
- 135 in the last 30 days · 13K all time · 1 likes
- Published
- 2024-12-11 · updated 2024-12-11 (1y ago)
- Languages
- en
- Tags
- llama, Llama-3, instruct, finetune, chatml, gpt4, synthetic data, distillation, function calling, json mode, axolotl, roleplaying
Other MLX builds of NousResearch/Hermes-3-Llama-3.2-3B
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| Hermes-3-Llama-3.2-3B-4bit (this one) | 4-bit | ≈ 1.8 GB | 8 GB+ | 135 |
| Hermes-3-Llama-3.2-3B-8bit | 8-bit | ≈ 3.4 GB | 8 GB+ | 150 |
| Hermes-3-Llama-3.2-3B-bf16 | bf16 | ≈ 6.4 GB | 16 GB+ | 35 |