5-bit Open on Hugging Face
Parameters
33B
Weights (est.)
≈ 23.0 GB
Mac memory (est.)
32 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-2.1-5bit --prompt "Hello"
Details
- Base model
- poolside/Laguna-XS-2.1 (quantized)
- Vendor
- poolside
- Task
- Text generation
- Quantisation
- 5-bit
- Licence
- openmdw-1.1
- Downloads
- 78 in the last 30 days · 936 all time · 2 likes
- Published
- 2026-07-03 · updated 2026-07-03 (4mo ago)
- Tags
- laguna, quantized, custom_code, eval-results
Other MLX builds of poolside/Laguna-XS-2.1
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| Laguna-XS-2.1-3bit | 3-bit | ≈ 14.6 GB | 24 GB+ | 109 |
| Laguna-XS-2.1-4bit | 4-bit | ≈ 18.8 GB | 24 GB+ | 2.1K |
| Laguna-XS-2.1-OptiQ-4bit | 4-bit | ≈ 18.8 GB | 24 GB+ | 371 |
| Laguna-XS-2.1-5bit (this one) | 5-bit | ≈ 23.0 GB | 32 GB+ | 78 |
| Laguna-XS-2.1-6bit | 6-bit | ≈ 27.2 GB | 48 GB+ | 117 |
| Laguna-XS-2.1-8bit | 8-bit | ≈ 35.5 GB | 48 GB+ | 189 |
| Laguna-XS-2.1-bf16 | bf16 | ≈ 66.9 GB | 96 GB+ | 79 |