overview of Supervised Fine-Tuning (SFT) for large language models, explaining it as a method to specialize pre-trained models for particular tasks by training them on curated, labeled datasets. It compares full fine-tuning with more efficient Parameter-Efficient Fine-Tuning (PEFT) methods like LoRA, highlighting their trade-offs. The text then outlines practical workflows for fine-tuning both API-based and open-weight models, emphasizing the critical importance of data quality and curation. Furthermore, it examines advanced alignment techniques, positioning SFT as a foundational step for methods such as Direct Preference Optimization (DPO), and discusses essential hyperparameters and evaluation metrics. Finally, the source addresses significant risks and limitations of SFT, including catastrophic forgetting and increased hallucination, and provides strategic recommendations for its effective application in real-world scenarios.