← Discover
Discover · GitHub · Llama.cpp↑ +158 today

Run AI models on your computer

Llama.cpp is a C and C++ based inference library that enables large language models to be run efficiently on standard hardware. The software makes it easier to run artificial intelligence models on local devices with its low memory usage and high performance-oriented structure.

Updates

  • September 15, 2026: Stars 127,101 → 128,278, latest release v0.4.1 (September 14, 2026).
  • September 5, 2026: Stars 125,883 → 127,101, latest release v0.4.0 (September 4, 2026).
  • August 27, 2026: Stars 125,068 → 125,883, latest release v0.3.0 (August 25, 2026).
  • August 22, 2026: Stars 124,998 → 125,068, latest release v0.2.0 (August 21, 2026).

What you get

  • It runs large language models efficiently on standard hardware.
  • Provides high performance with low memory usage.
  • Makes it easy to run AI models on local devices.

Installation

Clone source code
git clone https://github.com/ggml-org/llama.cpp.git
cd llama.cpp

Running it

Run GGUF model
llama-cli -m my_model.gguf

If you don't write code

🤖 Paste this into your AI agent (Claude Code · Codex · Antigravity)

I want to run large language models locally on my computer. How can I run my own model file using the llama.cpp tool with the command 'llama-cli -m my_model.gguf' or download and launch a model directly from Hugging Face with the command 'llama-cli -hf ggml-org/gemma-3-1b-it-GGUF'? Can you help me with the hardware requirements or model selection that I should pay attention to in this process?

Related dictionary terms

Who it is forIt is for users who want to run large language models on their own hardware without being dependent on the internet.
LicenseMIT

Links

TreScout did not build this tool · we found it in GitHub trends and wrote it up. This page describes the repository as of 2026-06-08: The star count and our text belong to that day, the repository may have changed since. Check the repository link for the current state. This page was machine-translated from the Turkish original · the Turkish version prevails.