Post-Training Quantization (PTQ)
An offline compression technique that converts weights to lower precision after model training finishes.
PTQ converts model weights (e.g. from FP16 to INT8) after training is complete. It requires no retraining and is a popular method to reduce model footprints for edge deployments.
Historical figures and technical concepts for informational purposes only. Not technical, professional, legal, or financial advice. Sources: Official Documentation.