* [llm-server v3.0.0](https://github.com/raketenkater/llm-server) – Hardware-detecting launcher that tunes and starts GGUF inference servers with automatic GPU placement, backend selection, and OpenAI-compatible serving.