PyPI page
Home page
Author:
Denis Ineshin
Summary:
Measure quantization quality loss on Apple Silicon MLX — KL divergence, top-token flip rate and perplexity delta for KV-cache and weight quantization
Latest version:
0.10.0
Required dependencies:
huggingface-hub
|
mlx
|
mlx-lm
|
numpy
|
pyarrow
Downloads last day:
4
Downloads last week:
35
Downloads last month:
386
7-day smoothing