4-bit Open on Hugging Face
Parameters
3.9B
Weights (est.)
≈ 2.2 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/Nanbeige4.1-3B-heretic-4bit --prompt "Hello"
Details
- Base model
- Daizee/Nanbeige4.1-3B-heretic (quantized)
- Vendor
- Daizee
- Task
- Text generation
- Quantisation
- 4-bit
- Licence
- apache-2.0
- Downloads
- 34 in the last 30 days · 472 all time · 0 likes
- Published
- 2026-02-16 · updated 2026-02-16 (8mo ago)
- Languages
- en, zh
- Tags
- llama, nanbeige, heretic, uncensored, decensored, abliterated
Other MLX builds of Daizee/Nanbeige4.1-3B-heretic
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| Nanbeige4.1-3B-heretic-4bit (this one) | 4-bit | ≈ 2.2 GB | 8 GB+ | 34 |
| Nanbeige4.1-3B-heretic-8bit | 8-bit | ≈ 4.2 GB | 8 GB+ | 43 |