⬢github C++ · 134.9K ★ +628 since we first saw it · pushed 4 min ago · MIT
ggml-org/llama.cpp
LLM inference in C/C++
llama.cpp runs large language model inference locally in plain C/C++ using the ggml tensor library, with no heavy dependencies. It supports running GGUF models from Hugging Face via a CLI, an OpenAI-compatible REST server with a built-in web UI, Docker images, and prebuilt binaries, enabling LLMs to run on consumer hardware.
Why now: not clear from the sources
Who it is for: Developers and hobbyists who want to run or serve LLMs locally on their own hardware.
ggml
Stars over our 129 snapshots: 134.2K to 134.9K, since 1 d ago.
Where people talked about it
- ⬢github ggml-org/llama.cpp 8 h ago
API: https://socialmediatrends-api.osmike.com/v1/repos/ggml-org/llama.cpp