4-bit Open on Hugging Face
Parameters
3.8B
Weights (est.)
≈ 2.1 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/numind-NuExtract-1.5-MLX-4bit --prompt "Hello"
Details
- Base model
- numind/NuExtract-1.5 (quantized)
- Vendor
- numind
- Task
- Text generation
- Quantisation
- 4-bit
- Licence
- mit
- Downloads
- 51 in the last 30 days · 842 all time · 2 likes
- Published
- 2024-11-27 · updated 2024-11-27 (1y ago)
- Languages
- multilingual
- Tags
- phi3, custom_code
Other MLX builds of numind/NuExtract-1.5
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| numind-NuExtract-1.5-MLX-4bit (this one) | 4-bit | ≈ 2.1 GB | 8 GB+ | 51 |
| numind-NuExtract-1.5-MLX-8bit | 8-bit | ≈ 4.1 GB | 8 GB+ | 51 |