Glossary
Fine-Tuning
Further training an existing model on your own examples so it adopts a specific style, format, or task without you having to spell it out every time.
Fine-tuning takes a general model and continues training it on a curated set of your own examples. The result is a version that has internalized a particular tone, format, or narrow task, so you need less instruction in each prompt.
It is not a cure-all. Fine-tuning teaches behavior and style, but it does not reliably add fresh facts; for keeping answers current and grounded in your data, retrieval is usually the better tool, and the two are often combined.
How we use it
We reach for fine-tuning when a task needs a consistent format or voice at scale, and we measure it against simpler approaches first so you only pay for it when it earns its place.
Get in touch
Want to put this into practice?
If this concept is relevant to something you're building, a short note is the fastest way to get practical help.