Predicting Quantization Price for Selecting PTQ Configurations Before Deployment

Excerpt

arXiv:2609.28270v1 Announce Type: new Abstract: Weight-space post-training quantization (PTQ) must choose finite formats, granularities, quantizer families, transformations, and bits before the completed quantized model reveals its output-distribution drift. Existing PTQ methods predict important pieces of this degradation, including reconstruction error, Hessian sensitivity, transformation effects, and downstream loss, but these pieces are usually scored after fixing the quantization geometry o