Skip to content

Polaris-4B-Preview-bf16

mlx-community/Polaris-4B-Preview-bf16
bf16 Open on Hugging Face
Parameters
4.0B
Weights (est.)
≈ 8.0 GB
Mac memory (est.)
16 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/Polaris-4B-Preview-bf16 --prompt "Hello"

Details

Base model
POLARIS-Project/Polaris-4B-Preview (finetune)
Vendor
POLARIS-Project
Task
Text generation
Quantisation
bf16
Licence
apache-2.0
Downloads
20 in the last 30 days · 190 all time · 0 likes
Published
2025-07-03 · updated 2025-07-03 (1y ago)
Languages
en
Tags
qwen3

Other MLX builds of POLARIS-Project/Polaris-4B-Preview

Build Quantisation Weights Mac memory Downloads
Polaris-4B-Preview-4bit 4-bit ≈ 2.3 GB 8 GB+ 32
Polaris-4B-Preview-5bit 5-bit ≈ 2.8 GB 8 GB+ 28
Polaris-4B-Preview-6bit 6-bit ≈ 3.3 GB 8 GB+ 24
Polaris-4B-Preview-8bit 8-bit ≈ 4.3 GB 8 GB+ 25
Polaris-4B-Preview-bf16 (this one) bf16 ≈ 8.0 GB 16 GB+ 20