Skip to content

Laguna-S-2.1-oQ3e-fast

mlx-community/Laguna-S-2.1-oQ3e-fast
3-bit Open on Hugging Face
Parameters
118B
Weights (est.)
≈ 51.5 GB
Mac memory (est.)
64 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-S-2.1-oQ3e-fast --prompt "Hello"

Details

Base model
poolside/Laguna-S-2.1 (quantized)
Vendor
poolside
Task
Text generation
Quantisation
3-bit
Licence
openmdw-1.1
Downloads
105 in the last 30 days · 2.3K all time · 10 likes
Published
2026-07-23 · updated 2026-07-24 (3mo ago)
Tags
laguna, quantized, custom_code

Other MLX builds of poolside/Laguna-S-2.1

Build Quantisation Weights Mac memory Downloads
Laguna-S-2.1-oQ2e-fast 2-bit ≈ 36.8 GB 48 GB+ 283
Laguna-S-2.1-oQ2e 2-bit ≈ 36.8 GB 48 GB+ 149
Laguna-S-2.1-OptiQ-2bit 2-bit ≈ 36.8 GB 48 GB+ 137
Laguna-S-2.1-oQ3e-fast (this one) 3-bit ≈ 51.5 GB 64 GB+ 105
Laguna-S-2.1-oQ3e 3-bit ≈ 51.5 GB 64 GB+ 83
Laguna-S-2.1-oQ4e 4-bit ≈ 66.2 GB 96 GB+ 1.2K
Laguna-S-2.1-oQ4e-fast 4-bit ≈ 66.2 GB 96 GB+ 173
Laguna-S-2.1-oQ5e 5-bit ≈ 80.9 GB 96 GB+ 95
Laguna-S-2.1-oQ6e 6-bit ≈ 95.6 GB 128 GB+ 191