llama.cpp
Local LLM RunnersHigh-performance LLM inference in plain C/C++
Replaces OpenAI API
Self-hosted apps by ggml-org
2 self-hosted projects maintained by ggml-org on GitHub — combined community: 185,457 GitHub stars.
Portfolio at a glance
The 2 projects span 2 categories: local llm runners and self-hosted ai. llama.cpp is the flagship at 131,102 stars — High-performance LLM inference in plain C/C++. 2 of 2 are written primarily in C++, so one toolchain covers the whole portfolio. The oldest dates from 2022 and the newest from 2023. 1 publish a Docker deployment path.
| Projects | 2 |
|---|---|
| Combined stars | 185,457 |
| Hosted on | GitHub |
| Languages | C++ (2) |
| Licenses | MIT (2) |
| Setup difficulty | Medium (1), Hard (1) |
| Min. RAM range | 1,024–8,192 MB |
| First releases | 2022–2023 |