Quantization

Reducing numeric precision in model weights to run inference cheaper and faster.

Quantization — Reducing numeric precision in model weights to run inference cheaper and faster..