← All projects
Developer building blocks

ggml

A small, fast core library for running AI models efficiently on many kinds of computers.

What it does

ggml is a low-level library that does the heavy number-crunching AI models need. A "tensor" is just a large grid of numbers, and ggml works with these efficiently.

It is written in plain C/C++ with no other dependencies and runs on many types of chips and in web browsers. It supports "quantization", shrinking models so they fit on ordinary hardware.

It defines the GGUF file format for AI models and is closely tied to llama.cpp, where changes to the core library are developed.

Who it's for

Low-level programmers building tools that run AI models on local hardware.

Can I use it without coding?

Developers only. Has to be built from source code and used in C/C++ programs. New to this? Start here.

Can I run it?

Cost
FreeNothing to pay or sign up for.
API key
Not needed

Try it

Using Claude Code, Cursor or another AI coding assistant? Paste the prompt and it will walk you through the setup.New to this?

Before you install

  • Only download from the official GitHub page. Copies with similar names exist for many popular projects.
  • License (MIT): Very permissive: free to use, change and share, including for business.
  • More on staying safe

Where people are talking about it