piwheels

mlx-quant-fidelity

Measure MLX quantization quality loss — KL divergence, perplexity, top-token agreement for KV cache and weights

Installation

In a virtualenv (see these instructions if you need to create one):

pip3 install mlx-quant-fidelity

Dependencies

• pyarrow
• huggingface-hub
• numpy
• mlx-lm
• mlx

Releases

Version	Released	Bullseye Python 3.9	Bookworm Python 3.11	Trixie Python 3.13	Files
0.4.0	2026-07-04
mlx_quant_fidelity-0.4.0-py3-none-any.whl (47 KB)
0.3.1	2026-06-23
mlx_quant_fidelity-0.3.1-py3-none-any.whl (43 KB)
0.3.0	2026-06-18
mlx_quant_fidelity-0.3.0-py3-none-any.whl (43 KB)
0.2.0	2026-06-15
mlx_quant_fidelity-0.2.0-py3-none-any.whl (30 KB)
0.1.0	2026-06-14
mlx_quant_fidelity-0.1.0-py3-none-any.whl (23 KB)

Issues with this package?

• Search issues for this package
• Package or version missing? Open a new issue
• Something else? Open a new issue