Versione quantizzata per MLX di https://huggingface.co/Almawave/Velvet-14B usando mlx_lm.convert, occupa 7.93GB e funziona molto bene sui Mac Silicon con 16GB di RAM

Safetensors

Model size

2.2B params

Tensor type

FP16

U32

Inference Providers NEW

This model is not currently available via any of the supported Inference Providers.

The model cannot be deployed to the HF Inference API: The model has no library tag.

Model tree for fabiolecca/almawave-Velvet-14B-MLX

Base model

Quantized

(7)

this model