model-optimization

Use when reducing model size, improving inference speed, or deploying to edge devices — covers quantization, pruning, knowledge distillation, ONNX export, and TensorRT optimizationUse when ", " mentioned.