4-bit Open on Hugging Face
Parameters
33B
Weights (est.)
≈ 18.8 GB
Mac memory (est.)
24 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/Laguna-XS.2-4bit --prompt "Hello"
Details
- Base model
- poolside/Laguna-XS.2 (quantized)
- Vendor
- poolside
- Task
- Text generation
- Quantisation
- 4-bit
- Licence
- apache-2.0
- Downloads
- 197 in the last 30 days · 5.0K all time · 4 likes
- Published
- 2026-04-28 · updated 2026-04-28 (6mo ago)
- Tags
- laguna, laguna-xs.2, custom_code
Other MLX builds of poolside/Laguna-XS.2
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| Laguna-XS.2-2bit-M-DQ | 2-bit | ≈ 10.5 GB | 16 GB+ | 21 |
| Laguna-XS.2-4bit (this one) | 4-bit | ≈ 18.8 GB | 24 GB+ | 197 |
| Laguna-XS.2-nvfp4 | 4-bit | ≈ 18.8 GB | 24 GB+ | 42 |
| Laguna-XS.2-mxfp4 | MXFP4 | ≈ 18.8 GB | 24 GB+ | 31 |
| Laguna-XS.2-5bit | 5-bit | ≈ 23.0 GB | 32 GB+ | 28 |
| Laguna-XS.2-6bit | 6-bit | ≈ 27.2 GB | 48 GB+ | 40 |
| Laguna-XS.2-mxfp8 | MXFP8 | ≈ 35.5 GB | 48 GB+ | 49 |
| Laguna-XS.2-8bit | 8-bit | ≈ 35.5 GB | 48 GB+ | 47 |
| Laguna-XS.2-bf16 | bf16 | ≈ 66.9 GB | 96 GB+ | 26 |