grug v1.1 Qwen 3.8 27B for mlx-serve (4-bit)

An Apache-2.0 MLX affine quantization of ProCreations/grug-v1.1-qwen-3.8-27b, matching the base model's license.

This is a text-only pack; it does not include the source model's vision assets.

Size and quantization

Safetensors payload: 15.13 GB / 14.09 GiB. Affine 4-bit quantization uses group size 64; norms, scales, and biases remain BF16.

Benchmarks

Benchmarks pending a reproducible rerun.

Usage

Requires mlx-serve v26.10.2 or later.

mlx-serve serve --model pgmharikrishnan/grug-v1.1-qwen-3.8-27b-MLX-Serve-4bit