AI Support & Agents

Fine-Tuning

Definition

Fine-tuning is the process of further training a pre-trained language model on a smaller, task-specific dataset so it adapts to a particular domain, style, or behavior — updating the model's weights rather than just its instructions.

Also known as: model fine-tuningsupervised fine-tuning

How it works

You start from a foundation model that already understands language broadly, then continue training it on curated examples — such as your support conversations — so it internalizes patterns like your tone or domain terminology. This changes the model's parameters, unlike prompting or RAG, which don't.

Fine-tuning is powerful for shaping how a model responds, but it doesn't reliably teach the model current facts about your business — for that, retrieval is usually the better tool.

Why it matters for support

Teams weigh fine-tuning against RAG. Fine-tuning bakes in style and behavior but is slower to update and can't keep up with changing prices or policies; RAG keeps answers current by retrieving live content. Many production support systems lean on RAG and prompting first, and fine-tune only when consistent behavior demands it.

Frequently asked

What is the difference between fine-tuning and RAG?

Fine-tuning changes the model's weights to adapt its behavior and style; RAG retrieves current information at query time to ground answers. Use fine-tuning for how the model responds, RAG for keeping facts accurate and up to date.

Do I need to fine-tune a model for customer support?

Usually not first. Most support use cases are handled well with a strong base model plus RAG and good prompting. Fine-tuning is worth it mainly when you need very consistent tone or behavior that instructions alone can't achieve.

Related terms

Foundation Model

A foundation model is a large AI model trained on broad, general-purpose data at scale that can be adapted to many downstream tasks — through prompting, retrieval, or fine-tuning — rather than being built for one narrow job..

RAG vs Fine-Tuning

RAG vs fine-tuning is the choice between grounding an AI model in external knowledge at query time (retrieval-augmented generation) versus adjusting the model's own weights on your data (fine-tuning) — two different ways to make a general model useful for your specific needs..

Retrieval-Augmented Generation (RAG)

Retrieval-augmented generation (RAG) is a technique where an AI model retrieves relevant information from your own knowledge sources at query time and uses it to ground its answer, instead of relying only on what it memorized during training..

Large Language Model (LLM)

A large language model (LLM) is a neural network trained on vast amounts of text to predict and generate language, enabling it to understand questions, summarize, classify, and write human-like responses..

Prompt Engineering

Prompt engineering is the practice of designing and refining the instructions given to a language model to get accurate, relevant, and reliable outputs for a specific task..

Put these ideas to work

Macha is an AI agent layer that sits on top of the help desk you already run — Zendesk, Freshdesk, Front, Intercom, or Gorgias.

Start Trial