mstyslavity
/

mamay-12b-mlx-optiq_4bit

Text Generation

4-bit precision

Model card Files Files and versions

From 44GB (full precision) to 7GB, 4.5 bit quantized!

Downloads last month: 57

Safetensors

Model size

12B params

Tensor type

BF16

·

U32

·

MLX

Hardware compatibility

Log In to add your hardware

4-bit

Model tree for mstyslavity/mamay-12b-mlx-optiq_4bit

Base model

shportix/MamayLM-Gemma-3-12B-IT-W4A16

Quantized

(1)

this model