# What is GGUF?

> GPT-Generated Unified Format

It is a file format designed for large artificial intelligence models to run efficiently on personal computers.

## Overview
GGUF is a packaging method that allows models to move more smoothly between your computer's processor and graphics card. By keeping the model in a single file, it simplifies installation and allows you to use hardware resources more balanced. It has become the standard, especially for those who want to run large models on home hardware.

*Analogy: It is like a special compression bag that allows you to fit a giant library into a single suitcase and carry it with you.*

## How it works
When you download the model, you usually get a single file with the .gguf extension. You can start using it immediately by introducing this file to local artificial intelligence running tools. You do not need to deal with extra installations or complex software configurations.

## Where it is used
It is used in local artificial intelligence applications, especially in tools like LM Studio or Ollama.

## Commonly confused with
It can be confused with general file formats or just compression methods.

## Frequently asked questions
**Why should I use GGUF?**
It allows you to easily manage the size and speed of the model according to your computer's power.

**Does every model work with GGUF?**
Yes, GGUF versions of most popular models are prepared and shared by the community.


## Related terms
- [Quantization](/en/dictionary/quantization/)
- [LLM](/en/dictionary/llm/)
- [Inference](/en/dictionary/inference/)

## Related tools
- [Llama.cpp](/en/discover/llama-cpp/)
- [Transcribe.cpp](/en/discover/transcribe-cpp/)

---
Source: TreScout Dictionary · https://trescout.com/en/dictionary/gguf/