What is Fine-tuning?
Fine-tuning is retraining an existing AI model on your own examples so it adopts your style, format, or domain knowledge.
Fine-tuning takes a general-purpose model and trains it further on a curated set of your own examples, so it behaves the way you need by default. If your support team answers tickets in a very specific tone, or your image style must stay consistent across a game or brand, fine-tuning bakes that pattern in instead of re-explaining it in every prompt. It is not a way to give a model fresh facts — that job belongs to retrieval-augmented generation, which looks documents up at answer time. Fine-tuning changes behaviour and style; retrieval changes knowledge. For most teams comparing tools, fine-tuning is a later-stage feature: you need a few hundred good examples, a way to evaluate the result, and a budget for retraining when the base model updates. Many products now offer a lighter version, marketed as custom models, brand voices, or trained styles, which is enough for the majority of marketing and design workflows.
Also known as
- model fine-tuning
- custom model training
Related tools
- Leonardo.ai — Generate game assets, concept art, and design visuals with AI.
- Stable Diffusion — Free, open-source image generation you can run on your own hardware.
- Jasper — AI copywriter built for marketing teams with brand voice control.
- Claude — Thoughtful, safety-focused AI assistant with excellent reasoning.
- ChatGPT — The most versatile AI assistant for writing, research, and brainstorming.
Related terms
- Large Language Model (LLM) — An LLM is an AI model trained on huge amounts of text that predicts language well enough to write, summarise, translate, and answer questions.
- RAG (Retrieval-Augmented Generation) — RAG is a technique where an AI tool searches your documents first, then writes an answer grounded in what it found.
- Generative AI — Generative AI is any AI that creates new content — text, images, audio, video, or code — rather than only analysing existing data.
- Open Source (AI Tools) — Open-source AI means the code, and sometimes the model weights, are published so anyone can inspect, self-host, and modify the software.
Category: AI Concepts