Word of the Day
Fine-Tuning
Fine-tuning further trains a model on your examples to permanently shift its style or behavior — heavier than prompting or RAG.
Get it freeFine-tuning takes a pretrained model and trains it further on a curated set of your own examples, permanently adjusting its weights so it defaults to your style, format, or task without being told each time. It's powerful for consistent tone or specialized formats at scale.
It is also the heaviest option: it needs quality example data, compute, and re-doing when things change. For most needs, a good prompt (with few-shot examples) or RAG (for fresh facts) is cheaper and more flexible. Reach for fine-tuning when prompting can't get you consistent enough behavior across many calls.
Related: rag · embeddings · few shot prompting · prompt engineering
One of these every weekday
Three minutes. One prompt, one word, one skill. You're now better at AI than you were yesterday.
Free. All terms