convert-to-quant

PyPI page
Home page
Author: None
License: MIT
Summary: Convert safetensors weights to quantized formats (FP8, INT8) with learned rounding optimization
Latest version: 1.2.0
Required dependencies: prodigy-plus-schedule-free
Optional dependencies: prodigy-plus-schedule-free | pytest | triton | triton-windows

Downloads last day: 61
Downloads last week: 139
Downloads last month: 436