# LLM Large Language Model. For example, ChatGPT, Claude. # LoRA A parameter efficient training method that typically keeps the base model’s weights frozen and trains a small set of added low-rank adapter weights (in 16-bit precision). # QLoRA Combines LoRA with 4-bit precision to handle very large models with minimal resources. # Reinforcement Learning (RL) Used when you need a model to excel at a specific behavior (e.g., tool-calling) using an environment and reward function rather than labeled data.