6-bit Open on Hugging Face
Parameters
3.8B
Weights (est.)
≈ 3.1 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/Phi-4-mini-reasoning-6bit --prompt "Hello"
Details
- Base model
- microsoft/Phi-4-mini-reasoning (quantized) · Phi 4 mini reasoning on the radar →
- Vendor
- Microsoft · Phi
- Task
- Text generation
- Quantisation
- 6-bit
- Licence
- mit
- Downloads
- 68 in the last 30 days · 459 all time · 1 likes
- Published
- 2025-05-01 · updated 2025-05-01 (1y ago)
- Languages
- en
- Tags
- phi3, math, code
Other MLX builds of microsoft/Phi-4-mini-reasoning
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| Phi-4-mini-reasoning-4bit | 4-bit | ≈ 2.2 GB | 8 GB+ | 286 |
| Phi-4-mini-reasoning-6bit (this one) | 6-bit | ≈ 3.1 GB | 8 GB+ | 68 |
| Phi-4-mini-reasoning-8bit | 8-bit | ≈ 4.1 GB | 8 GB+ | 85 |
| Phi-4-mini-reasoning-bf16 | bf16 | ≈ 7.7 GB | 16 GB+ | 77 |