Draft page

This page is an outline. It describes what will be covered and is not yet complete technical documentation.

Quantization stores and computes with lower-precision numbers, reducing memory footprint and often increasing throughput. The saving is real and frequently large; the accuracy cost is real too, and workload-dependent.

This page explains the main approaches and how to decide whether a given quantization is acceptable for your task.

What you will learn

  • What quantization changes in the memory budget and the compute path.
  • How post-training quantization differs from quantization-aware training.
  • Which parts of a model are most sensitive to reduced precision.
  • How to measure quality loss on your own task rather than trusting a general claim.

This page is an outline. The subsections below are the planned structure; they are filled in as the handbook is written.

Precision and numeric formats

Draft

Post-training quantization

Draft

Quantization-aware training

Draft

Weight-only versus activation quantization

Draft

Measuring the accuracy impact

Draft

Interaction with other optimizations

Draft