▲ 618 today135k starsC++
llama.cpp lets anyone run LLMs locally with no dependencies
What is llama.cpp?
llama.cpp is a C/C++ library that runs large language models on CPUs, GPUs and even phones without installing heavy frameworks. It supports quantized models and offers an OpenAI-compatible server.
Who it is for
Developers and teams who want fast, cheap inference on their own hardware can spin up a local model without cloud costs.
Try it in 60 seconds
You need: A Unix-like terminal with curl
curl -LsSf https://llama.app/install.sh | shllama cli -hf ggml-org/Qwen3.5-0.8B-GGUFllama serve -hf ggml-org/Qwen3.5-0.8B-GGUF
MIT licenseUpdated todayC++2.6k open issues
New projects are not vetted. Skim the code, and run it in a fresh folder or a container before giving it your files or keys.
The story so far
Early Signal first spotted ggml-org/llama.cpp on November 12, 2026, when it had 134k GitHub stars. It now has 135k. On its busiest day it gained 618 stars. It is published under the MIT license and has 25.7k forks. Early Signal files it under LLMs, Inference, Coding.
Questions about llama.cpp
- Is llama.cpp free and open source?
- Yes. ggml-org/llama.cpp is open source under the MIT license, and the code is on GitHub at https://github.com/ggml-org/llama.cpp.
- How do I install llama.cpp?
- Use one of these, depending on your setup: curl -LsSf https://llama.app/install.sh | sh · llama cli -hf ggml-org/Qwen3.5-0.8B-GGUF · llama serve -hf ggml-org/Qwen3.5-0.8B-GGUF
- What do I need to run llama.cpp?
- A Unix-like terminal with curl