8-bit Open on Hugging Face
Parameters
850M
Weights (est.)
≈ 1.2 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/bitnet-b1.58-2B-4T --prompt "Hello"
Details
- Base model
- microsoft/bitnet-b1.58-2B-4T (quantized)
- Vendor
- Microsoft · Phi
- Task
- Text generation
- Quantisation
- 8-bit
- Licence
- mit
- Downloads
- 265 in the last 30 days · 3.7K all time · 7 likes
- Published
- 2025-06-10 · updated 2025-06-10 (1y ago)
- Languages
- en
- Tags
- bitnet, chat, large-language-model, custom_code
Other MLX builds of microsoft/bitnet-b1.58-2B-4T
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| bitnet-b1.58-2B-4T-6bit | 6-bit | ≈ 495 MB | 8 GB+ | 39 |
| bitnet-b1.58-2B-4T-4bit | 4-bit | ≈ 631 MB | 8 GB+ | 165 |
| bitnet-b1.58-2B-4T-8bit | 8-bit | ≈ 662 MB | 8 GB+ | 65 |
| bitnet-b1.58-2B-4T (this one) | 8-bit | ≈ 1.2 GB | 8 GB+ | 265 |