Fine-Tuning: Full, LoRA, QLoRA

Moving behavior into the weights: full fine-tuning, low-rank adapters, quantized training — and the recipe that keeps it honest.

Content last verified 2026-09.

Lessons

1. What Fine-Tuning Changes
2. LoRA: Low-Rank Adaptation
3. QLoRA: Training on a Budget
4. The Practical Recipe

Sources

• Hu et al. (2021) — LoRA: Low-Rank Adaptation of Large Language Models
• Dettmers et al. (2023) — QLoRA: Efficient Finetuning of Quantized LLMs
• Hugging Face PEFT documentation — parameter-efficient fine-tuning methods