BTC
ETH
SOL
XRP
BNB
TON
DOT
LTC
POL
HBAR
NEAR
ARB
UNI
Training an existing AI model a little more on your own examples, so it gets better at one task or speaks in one style.
AI Drafted with AI help · Reviewed by Keshav Aggarwal · Updated
In simple words
A chef who can cook everything joins your restaurant. You spend a week showing him your recipes. Now he cooks your dishes perfectly.
That extra week of training is fine-tuning.
The technical side
Fine-tuning continues training a pre-trained model on a smaller, task-specific dataset, adjusting its weights. Full fine-tuning updates every parameter; cheaper methods such as LoRA train a small set of extra weights instead. Many providers sell fine-tuning through their APIs, and open-weights models can be fine-tuned on your own GPUs.
For adding facts that change often, RAG is usually cheaper and easier to keep up to date than fine-tuning.
Fine-Tuning kya hai?
Fine-tuning matlab pehle se bane AI model ko apne examples se thoda aur sikhana, taaki wo ek kaam mein expert ban jaaye.
Related terms
- LLM (Large Language Model)AIAn AI model trained on huge amounts of text so it can read and write language, like the models behind ChatGPT, Claude and Gemini.
- Open WeightsAIAn AI model whose trained numbers (weights) are published, so anyone can download it and run it on their own computers.
- ParametersAIThe numbers inside an AI model that it learned in training. Model size is counted in parameters, often billions of them.