Quantize

Compress models without losing accuracy

Quantize applies pruning, distillation, and quantization to shrink models for edge and cost-sensitive deployments.