Overview
@xiciyang/inference-quantization
Quantize a model checkpoint and prove it: choose a method (default AWQ+INT4), PTQ-calibrate with a leak-free split, pass an independent accuracy gate and a performance gate through real serving endpoints, auto-degrade on failure, and sign off or roll back — fully audited.
README / EN
Package documentation
Registry summary
Quantize a model checkpoint and prove it: choose a method (default AWQ+INT4), PTQ-calibrate with a leak-free split, pass an independent accuracy gate and a performance gate through real serving endpoints, auto-degrade on failure, and sign off or roll back — fully audited.
dsh.pub verifies the pinned bundle contract, runtime facts, and distribution semantics. The complete README remains in the source repository.
Read the full README on GitHub