Code for paper "Adversarial Moment-Matching Knowledge Distillation for Large Language Models" in NeurIPS'2024.