What is model quantization in production ML?

Quantization reduces model size by using lower precision weights.

It converts float32 weights to int8 or float16, reducing memory and improving inference speed. It is widely used in edge deployment and mobile ML systems.